𝕐 on Ray Yang, Ph.D.

Bitcoin Value

Sun, 04 Feb 2024 00:00:00 +0000

In a world where monkeys from the Web2 jungle are perplexed by humans trading tokens for bananas in the Web3 era, the debate around Bitcoin’s value mirrors a similar situation in human history. There exists a gap between the fading world of traditional finance and the pioneering world of digital assets.

Upside

Bitcoin’s Value as a Decentralized Asset: One of the fundamental appeals of Bitcoin is its decentralization. Unlike fiat currencies, which are controlled and issued by central banks, Bitcoin operates on a decentralized network. This decentralization is viewed as a strength by many, as it reduces reliance on central authorities and is resistant to censorship and manipulation. This aspect offers a unique value proposition, especially in countries with unstable economies or where traditional banking systems are not accessible to everyone.
Store of Value as Digital Gold: Bitcoin, often compared to gold for its value as a store of wealth and a scarce commodity—there will only ever be 21 million bitcoins. This scarcity helps protect against inflation, similar to gold. However, Bitcoin’s true value lies beyond physical attributes; it transcends borders and barriers, serving as an exchange medium connecting disparate demand and supply. Though lacking traditional intrinsic value, Bitcoin’s ability to operate globally without the constraints of physical borders or disruptions enhances its appeal as “digital gold.”
Technological Innovation and Network Effects: The technology underlying Bitcoin, blockchain, is a significant innovation. It offers a secure, transparent, and immutable way of recording transactions. This technology has the potential to revolutionize not just currency, but various sectors like supply chain management, voting systems, and digital identities. The value of Bitcoin, therefore, is also tied to the innovative potential of its underlying technology and the possibilities it opens up beyond being just a currency. As long as people find value in its properties and network, its market value persists.

Flip-side

Lack of Intrinsic Value: Unlike tangible assets like food, which can be eaten, or artwork and jewelry, which can be visually appreciated and worn, Bitcoin does not possess such intrinsic physical qualities. It cannot be utilized in a physical sense. For instance, while you can display a piece of artwork in your home or drive a car, Bitcoin does not offer any physical utility or aesthetic pleasure. It exists purely as a digital entity without any physical form or practical use in the everyday world.
Non-Redeemability: Traditional financial instruments like fiat currencies, shares, or gift cards have an issuing authority that provides a form of redemption. For example, a casino chip can be exchanged for cash, and fiat currencies are backed by governments that guarantee their value for transactions and debt servicing. In contrast, Bitcoin lacks such an issuing authority willing to redeem it. Satoshi Nakamoto, the creator of Bitcoin, does not offer any form of redemption for Bitcoins, making them fundamentally different from these traditional financial instruments.
Limited Utility: Bitcoin’s utility is limited compared to traditional assets or currencies. For example, you can use fiat currency to buy goods and services almost anywhere, and shares in a company can be redeemed for a share of its assets if it liquidates. Bitcoin, however, cannot be used in such a direct way. It’s not universally accepted for transactions, nor does it provide the holder with a claim on physical assets or services. Its role is more akin to a number in an online game, where its value is not tied to a physical commodity or a guaranteed service, but rather to the collective agreement and interest of its users.

the Success Formula for Entrepreneurs

Wed, 01 Dec 2021 00:00:00 +0000

the Formula

$y = f(x) + \epsilon$

the Four Elements

$y$: success level (the dependent variable): something you can keep in mind but is NOT under your control. It’s the outcome that “depends” on something else (on the right-hand side of the formula).

$x$: the input (the independent variable): something you can control.

$f$: your method (the functional form): something that determines how your input converts into your outcome.

$\epsilon$: noise (the error term): good luck and bad luck that are totally inevitable and completely out of your control. But it is very important so you should take notice and be mindful of it.

How to Use the Formula

setting goal

Set and Forget. An entrepreneur should first model his/her $y$ (success level). Set a goal on $y$ but learn to shift focus to $f(x)$. By making the shift, it is important to be emotionally detached from $y$ when working on $f(x)$. This is because if you are so emotionally fixated on $y$, you will forget $f(x)$. Then your formula will become merely “$y = \epsilon$”, which means your outcome will totally depend on luck. In this case, you will learn nothing and control nothing.

Challenging, Specific, and Multi-level. Setting a fixed, challenging goal ($y$) helps you search for a better $f$. The right way is to be fixed on your $y$ but be flexible on your $f$, because $y$ (what you will achieve) depends on $f(x)$ (your method applied to your input), but not the other way around. Also, write down your goals at multiple levels. Don’t fall into the “security trap.” If you only have a “capped”, single-level goal for success, your learning and exploration for better $f$ will slow down or even stop as soon as you hit your first-level goal.

shifting focus

Understand the “$=$”: Goal is NOT Method. $f(x)$ is the working process of how you use a given “method” $f$ to transform your input $x$, e.g., how you structure your schedule, how you organize your resources, how you utilize your skills, and, in the most general term, your “way of thinking.” When you understand this part, you will never mess up your goal $y$ with your method $f$ and your input $x$.

Avoid “Short-term” Traps. Whenever you feel pressures from short-term $y$ or are eager to receive instant gratifications from short-term $y$, your should try to get your focus back on $f(x)$. $f(x)$ is under your control “right here and right now,” $y$ is not. You won’t be successful by focusing on success itself ($y$). What you should do is to learn this simple trick: “do the right things; success will follow.” Pressures and temptations from short-term $y$ can only take your productivity (your creativity and your concentration) away from you. Moreover, the transformation process of $y = f(x)$ typically takes time to present itself. In the short term, you only tend to see $\epsilon$.

Develop a Skill of Focus-shifting:

When tasking on $x$, you should concentrate at “right here and right now” and throw $y$ completely into the back of your mind.
When taking a break from $x$, revisit your $y$ and analyze $\epsilon$ to improve your $f$.
When implementing on $f(x)$, use schedules to develop routines for making plans and doing reflections, and let a well-paced, non-stopping process to carry you through.

developing advantage

Accept and Learn. The best part is that your outcome depends on how you deal with $\epsilon$, be it flukes or failures. You need to do two things. First, remember that $\epsilon$ is not the whole part of $y$. Otherwise, you will give up easily after encountering failures. The secret behind most success is that “winners never quit.” Second, learn from it. Understand that different people have different versions of the formula. Your $\epsilon$ can also be other people’s $\epsilon$. But if you can carve out a portion of $\epsilon$ and put it into your $f(x)$, which is a trick you figure out before others, or something you know but others don’t, you will have an unfair advantage. In this case, the $\epsilon$ turns around to help you, by setting up a barrier to prevent others from stealing your success.

Stone to Gold. How to do it? Analyze $\epsilon$ carefully, and treat it as the “feedback” on your new $f$ (method) and different combinations of $x$ (input), you will develop a refined $f(x)$ with private knowledge to create and sustain your advantage. Now, your $f(x)$ (method on input) is others’ “luck and miracles” $\epsilon$. The most amazing thing is that only you know how to use a reliable system $f(x)$ to mass-produce “luck and miracles” in the eyes of others, and only you know there is no such things as “luck and miracles.”

Own your Errors. $\epsilon$ has another nice name – error. Most successful people crack the code from trials and errors. Learning from feedback is key. The progress of learning is a function of the number of iterations $x$. A greater number of iterations can generate more helpful data for us to learn from. But sometimes the feedback can be quite subtle. Without smoking-gun evidence on whether your method $f$ will work, we have to rely on our own judgment, a leap of faith, or risk-taking. When you feel it’s too hard, it might be worth holding on to your old way $f$, or it might be a good time to change your method $f'$. But never quit too soon, because the path to success is typically non-linear, so it is often critical to hold on until your hit the “critical mass” for any good methods $f$ to work.

Plug Your Life into the Formula

$y_{t} = f(x_{t-1}) + \epsilon$ ($\text{your present outcome = your old method applied to your old input + luck}$)

$y_{t+1} = f'(x_t) + \epsilon$ ($\text{your future outcome = your new method applied to your new input + luck}$)

Only by taking full responsibility of your present outcome $y_{t}$ as a result of your old method $f(x_{t-1})$ and setting a new goal $y_{t+1}$ can you feel a strong desire for new method $f'(x)$.
Take control of your new method $f'(x)$, by preparing yourself to learn and master $f'$ with inevitable $\epsilon$.
Take actions, and most importantly, hold on to process in $x_t$ that will guarantee your future success $y_{t+1}$.

the Backend Story

An entrepreneur friend asked me for something that is worth hearing. This is what I told him––a regression model.

A List of Data Sources for Real Estate Market Analysis

Tue, 09 Nov 2021 00:00:00 +0000

General Market / Environmental Condition

Price

Supply

U.S. Census Bureau on New Residential Construction

Buyer/Renter Demographics

Neighborhood / Local Information

Online Listing

Property Information

CRS Data

Calculators

Regression Discontinuity Design

Wed, 23 Jun 2021 00:00:00 +0000

the Causality Graph

Remove the Differences in Y NOT explained by Treatment and Time

Difference-in-Difference

Wed, 16 Jun 2021 00:00:00 +0000

the Causality Graph

Remove the Differences in Y NOT explained by Treatment and Time

Matching

Wed, 09 Jun 2021 00:00:00 +0000

the Causality Graph

Removing the “Differences in Y explained by X” (to find out the pure effect of Treatment on Y)

Fixed Effects

Wed, 02 Jun 2021 00:00:00 +0000

the Causality Graph

Remove “Differences in X explained by Person” vs. Remove “Differences in Y explained by Person”

Remove Persons’s influence on “the effect of X on Y” step by step

Play the Thinking Game -- from Seeing from Eyes to Seeing with the Mind

Wed, 02 Jun 2021 00:00:00 +0000

The ontological and epistemological underpinning of research methodology.

Ontology: The truth lies in the things that we can’t see from our eyes. The invisible generates the visible under natural laws and processes. Thus, the “observational” data (what we see from our eyes) stems from the underlying mechanism (what we can’t see from our eyes). Then, our mind comes to help (to develop theories), with reasoning and inference, see what is behind the data.

Epistemology: We need to be suspicious about what we see from our eyes because data can be misguiding, especially when we shut off our reasoning mind and stop making the inference. In other words, the frontend phenomenon (e.g., the data) can be illusional when we stop seeking to understand the backend processes (e.g., the processes that generate the data). Researchers need to be self-reflective and self-critical about HOW we know what we know.

Inference: We can use what we see from our eyes (observational data) to recover the underlying truth with our mind (reasoning techniques). There are two elements: first, we know that what we see from eyes (e.g., data) comes from the underlying truth (e.g., probability); second, we know some properties about the truth (e.g., distribution) so that we generate estimates from the what our eyes see (e.g., sample).

Counterfactuals: The reality (i.e., a data point), events that have already occurred, is a probabilistic ‘realization’ of the underlying mechanism (i.e., probability distribution). To recover the truth, we need to construct the alternatives to the reality just like looking for the missing pieces to complete a puzzle. Fact (what happens) does not equate to the truth (which determines what happens). Fact itself does not predict fact. Truth, recovered from the fact and its alternative, will be able to predict fact.

Post-treatment Control

Wed, 26 May 2021 00:00:00 +0000

the Causality Graph

Remove “Differences in X explained by C” vs. Remove “Differences in Y explained by C”

Remove C’s influence on X and Y step by step

Collider Bias

Wed, 19 May 2021 00:00:00 +0000

the Causality Graph

Remove “Differences in X explained by C” vs. Remove “Differences in Y explained by C”

What the model does: Remove C’s influence on X and Y step by step

Instrument Variable

Wed, 12 May 2021 00:00:00 +0000

Causality Graph

Generate the Data for Analysis

# DGP
set.seed(66)
# Z is a binary variable in this case, but it can also be continuous.
df <- data.frame(Z = c(rep(0, 100), rep(1, 100)), 
                 W = rnorm(200)) %>%
  # Z affects X
  mutate(X = .5 + 2*W + 2*Z + rnorm(200)) %>% 
  # Z does NOT affect either Y or W (God's Game)
  mutate(Y = -X + 4*W + 1 + rnorm(200)) %>% 
  group_by(Z) %>%
  mutate(mean_X=mean(X), mean_Y=mean(Y)) %>%
  ungroup()

Visualization

Remove “Differences in X NOT explained by Z” vs. Remove “Differences in Y NOT explained by Z”

Retain “Differences in X and Y explained by Z” step by step

Run the Regression and Check the Estimate of X

library(AER)
# Z is the instrument variable
summary(ivreg(Y ~ X | Z , data = df)) %>% 
  coef() %>% 
  regrrr::to_long_tab()

##   n.r        var_      beta
## 1   1 (Intercept)    1.318*
## 2   1               (0.583)
## 3   2           X -1.089***
## 4   2               (0.300)

Sample Selection Bias

Wed, 12 May 2021 00:00:00 +0000

Causality Graph

DGP

set.seed(66)
# Generate Exogenous Variables
df <- data.frame(X = runif(300) * 10, 
                 W = runif(300) * 5, 
                 U = runif(300) * 2, 
                 Z = c(rep(0, 150), rep(1, 150))) %>%
  # God's Game
  # X (Visible) affects Y, W (Invisible) and U (Invisible) affect Y
  mutate(Y = .1 + .5*X + 0.8*W + 0.2*U + rnorm(300),
  # X (Visible) affects Y, W (Invisible) and U (Invisible) affect Latent Selection Function
  # Z (Visible) affect Latent Selection Function
         select_ = 10 + 3*X + 2*W + 1*U + .5*Z + rnorm(300),
  # Latent Selection Determines the True Selection
         select = ifelse(select_ > mean(select_), 1, 0)
         )

Run Heckman Regression

run_selection <- glm(select ~ X + Z, family = binomial( link = "probit" ), data = df)
df$IMR <- dnorm(run_selection$linear.predictors)/pnorm(run_selection$linear.predictors)
summary(run_selection)

## 
## Call:
## glm(formula = select ~ X + Z, family = binomial(link = "probit"), 
##     data = df)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -2.0059  -0.1287  -0.0013   0.1240   2.5195  
## 
## Coefficients:
##             Estimate Std. Error z value Pr(>|z|)    
## (Intercept)  -5.4514     0.6613  -8.244   <2e-16 ***
## X             1.0163     0.1213   8.380   <2e-16 ***
## Z             0.6183     0.2704   2.287   0.0222 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 415.03  on 299  degrees of freedom
## Residual deviance: 113.84  on 297  degrees of freedom
## AIC: 119.84
## 
## Number of Fisher Scoring iterations: 8

run_outcome <- lm(Y ~ IMR + X, data = df[which(df$select == 1),])
summary(run_outcome)

## 
## Call:
## lm(formula = Y ~ IMR + X, data = df[which(df$select == 1), ])
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.4067 -0.8600  0.0496  0.7680  3.4331 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)   2.4928     0.8875   2.809  0.00569 ** 
## IMR           0.9237     0.4398   2.100  0.03749 *  
## X             0.4922     0.1099   4.477 1.56e-05 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.338 on 139 degrees of freedom
## Multiple R-squared:  0.1532, Adjusted R-squared:  0.141 
## F-statistic: 12.58 on 2 and 139 DF,  p-value: 9.55e-06

Visualization

Double Check

# Run OLS (which is biased)
run_biased <- lm(Y ~ X, data = df[which(df$select == 1),])
summary(run_biased)

## 
## Call:
## lm(formula = Y ~ X, data = df[which(df$select == 1), ])
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.5643 -0.9695  0.1065  0.7926  3.2590 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)   4.0168     0.5172   7.766 1.55e-12 ***
## X             0.3114     0.0692   4.500 1.42e-05 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.354 on 140 degrees of freedom
## Multiple R-squared:  0.1263, Adjusted R-squared:  0.1201 
## F-statistic: 20.25 on 1 and 140 DF,  p-value: 1.419e-05

# Check the Inclusion Restriction
run_test_instrument <- lm(Y ~ X + Z, data = df[which(df$select == 1),])
summary(run_test_instrument)

## 
## Call:
## lm(formula = Y ~ X + Z, data = df[which(df$select == 1), ])
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.3369 -0.9958  0.1245  0.8445  3.1289 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  3.71761    0.54715   6.794 2.93e-10 ***
## X            0.32192    0.06913   4.657 7.42e-06 ***
## Z            0.37117    0.23158   1.603    0.111    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.347 on 139 degrees of freedom
## Multiple R-squared:  0.1422, Adjusted R-squared:  0.1299 
## F-statistic: 11.52 on 2 and 139 DF,  p-value: 2.347e-05

# In Reality, Only God Can Run This Regression (by Knowing about the Truth and Seeing the Invisibles).
run_ture <- lm(Y ~ U + W + X, data = df)
summary(run_ture)

## 
## Call:
## lm(formula = Y ~ U + W + X, data = df)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -3.02525 -0.64811 -0.01175  0.56035  2.71664 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept) 0.008025   0.188084   0.043  0.96600    
## U           0.330793   0.099258   3.333  0.00097 ***
## W           0.773520   0.040014  19.331  < 2e-16 ***
## X           0.499843   0.020288  24.637  < 2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.9544 on 296 degrees of freedom
## Multiple R-squared:  0.7703, Adjusted R-squared:  0.768 
## F-statistic: 330.9 on 3 and 296 DF,  p-value: < 2.2e-16

Control Variable

Wed, 05 May 2021 00:00:00 +0000

the Causality Graph

Simpson’s Paradox

Remove “Differences in X explained by W” vs. Remove “Differences in Y explained by W”

Remove W’s influence on X and Y step by step

Quant Finance Data Source (Highly Recommended)

Tue, 09 Jun 2020 00:00:00 +0000

Tushare

Tushare

Tushare is an open-source data interface for Chinese stock market, cryptocurrency, and alternative data. The data can be accessed through its Python module. It is the best data source of this kind I have seen so far.

the Positioning of Stock Fundamentals and Abnormal Stock Returns

Tue, 19 May 2020 00:00:00 +0000

1 Motivation

According to Warren Buffet, buying a stock is like “acquiring an ownership interest in the business of a company.” Buffet tends to only care about the fundamentals of the stocks he buys.

Here the question is: do companies have a “fundamental positioning” – a relative standing to peer firms in terms of the financial fundamentals — to determine the differences in stock returns?

As a strategist, I consider a stock as a winner only when it beats the market. In other words, I only care about the “winner premium” — the stock returns above the expected level (a.k.a. above-normal rate of returns). I will explore utilizing the “fundamental positionings” to pick (predict) winner stocks.

2 Pipeline

To implement this idea, I used XGBoost (an implementation of gradient boosted decision trees), which requires minimal feature engineering/selection, to predict the “buy-and-hold” abnormal returns using the fundamental positionings prior to the holding period (90 days). To examine the predictive power of fundamental positionings, I left the base financial ratio variables along with the fundamental positionings (the relative percentile rank within an industry based on these ratios) in the feature space. After finding the best model from bayesian hyperparameter tuning, we can then evaluate the model performance and see the relative importance of fundamental positioning features and the financial ratios.

The data is streamed from “ Tushare,” which provides free access to the SH/SZ stock exchange.

3 Result

In the above graph, the points in yellow are winners that beat the market, whereas the points in purple are “losers” that fail to beat the market. If we use the algorithm to pick the top three winners, I will indeed generate positive abnormal returns for us. The mean squared error of the model is 0.019.

This graph shows the relative importance of the features. I added a prefix “R” (relative standing) to the variable names to denote the “fundamental positioning” features. Around half of these important features are fundamental positionings. This shows the positioning (which is based on the percentile ranking of a focal company relative to industry peers) does contribute to the prediction of stock market returns, in addition to the financial ratios (which does not involve firm-to-firm comparisons).

An Ongoing Collection of Mission Statements and Vision Statements

Sun, 03 May 2020 00:00:00 +0000

Here is a collection of mission statements and vision statements.

Mission statements should clarify a company’s core ideology. Vision statements should indicate a company’s long-term goals.

Both mission and vision are the things that won’t change in the short-term, which serve the purpose of guiding behavior, inspiring stakeholders, and stimulating progress.

Mission Statements

3M: 3M is committed to actively contributing to sustainable development through environmental protection, social responsibility, and economic progress
Adidas: The Adidas Group strives to be the global leader in the sporting goods industry with brands built on a passion for sports and a sporting lifestyle. We are committed to continuously strengthening our brands and products to improve our competitive position.
Amazon: To offer our customers the lowest possible prices, the best available selection, and the utmost convenience.
Apple: To bringing the best user experience to its customers through its innovative hardware, software, and services.
Cisco: Shape the future of the Internet by creating unprecedented value and opportunity for our customers, employees, investors, and ecosystem partners.
Coca Cola: To refresh the world in mind, body, and spirit, to inspire moments of optimism and happiness through our brands and actions, and to create value and make a difference.
Costco: To continually provide our members with quality goods and services at the lowest possible prices.
Disney: To entertain, inform and inspire people around the globe through the power of unparalleled storytelling, reflecting the iconic brands, creative minds and innovative technologies that make ours the world’s premier entertainment company.
Facebook: To give people the power to build community and bring the world closer together.
Google: To organize the world’s information and make it universally accessible and useful.
Ikea: Offer a wide range of well-designed, functional home furnishing products at prices so low that as many people as possible will be able to afford them.
Intel: Utilize the power of Moore’s Law to bring smart, connected devices to every person on earth.
Microsoft: To empower every person and every organization on the planet to achieve more.
Netflix: We promise our customers stellar service, our suppliers a valuable partner, our investors the prospects of sustained profitable growth, and our employees the allure of huge impact.
Nike: Do everything possible to expand human potential
Southwest Airlines: Dedication to the highest quality of customer service delivered with a sense of warmth, friendliness, individual pride, and company spirit.
Starbucks: To inspire and nurture the human spirit – one person, one cup and one neighborhood at a time.
Tesla: To accelerate the world’s transition to sustainable energy.
Uber: To bring transportation — for everyone, everywhere.
Walmart: To save people money so they can live better.

Vision Statements

3M: 3M Technology Advancing Every Company; 3M products enhancing every home; 3M innovation improving every life.
Adidas: To be the design leaders with a focus on getting the best out of the athletes with performance guaranteed products in the sports market globally.
Amazon: To be Earth’s most customer-centric company, where customers can find and discover anything they might want to buy online.
Apple: We believe that we are on the face of the earth to make great products and that’s not changing.
Cisco: Changing the way we work, live, play, and learn.
Coca Cola: Inspiring each other to be the best we can be by providing a great place to work.
Costco: To provide a wide selection of merchandise, plus the convenience and exclusive member services, all designed to make the shopping experience pleasurable.
Disney: To be one of the world’s leading producers and providers of entertainment and information.”
Facebook: People use Facebook to stay connected with friends and family, to discover what’s going on in the world, and to share and express what matters to them.
Google: To provide access to the world’s information in one click.
Ikea: To create a better everyday life for the many people.
Intel: If it is smart and connected, it is best with Intel.
Microsoft: to help people and businesses throughout the world realize their full potential.
Netflix: Becoming the best global entertainment distribution service.
Nike: To bring inspiration and innovation to every athlete in the world.
Southwest Airlines: To become the world’s most loved, most flown, and most profitable airline.
Starbucks: To establish Starbucks as the premier purveyor of the finest coffee in the world while maintaining our uncompromising principles while we grow.
Tesla: To create the most compelling car company of the 21st century by driving the world’s transition to electric vehicles.
Uber: Smarter transportation with fewer cars and greater access. Transportation that’s safer, cheaper, and more reliable; transportation that creates more job opportunities and higher incomes for drivers.
Walmart: Be the destination for customers to save money, no matter how they want to shop.

Want to Add to the List?

Cognitive Biases to Avoid in Decision Making

Mon, 20 Apr 2020 00:00:00 +0000

Affect Heuristic
Letting current positive and negative emotions (e.g., fear, pleasure, surprise) influence decisions.
Ambiguity Aversion
Tendency to prefer known/certain probabilities over uncertain probabilities.
Anchoring Bias
Letting an initial piece of information influence decisions.
Availability Bias
Letting instances that came to mind at ease influence decisions.
Bandwagon Effect
Doing what others do.
Confirmation Bias
Tendency to filter information that confirms one’s prior beliefs.
Decoy Effect
Letting asymmetrically dominated alternatives influence choice.
Default Bias
The psychological inertia to remain at the status quo.
Framing Effect
Letting the way the options are framed influence decisions.
Impact Bias
Overestimate the long-term impact of an event.
Loss Aversion
Letting losses loom larger than corresponding gains.
Overconfidence
Overvalue the benefit and underestimate the risk.
Outcome Bias
Letting a prior decision outcome influence subsequent independent decisions.
Representativeness Bias
Tendency to compare a new event with a prototype in mind.
Selective Perception
Tndency to ignore information to avoid emotional discomfort and contradictpry beliefs.
Sunk-cost Bias
Tendency to continue an endeavor once an investment (money, effort, or time) has been made.

Tracking the Change of Industry Positioning Caused by an Acquisition

Sun, 19 Apr 2020 00:00:00 +0000

1. Motivation

Due to enduring resource heterogeneity, firms tend to hold persistent positionings within an established industry (like commercial banking). However, transformative corporate actions may alter a firm’s asset structure and create a new industry for the acquirer. Thus, we can map out the industry positions and track the position change caused by an acquisition.

2. Mapping the Positioning and Tracking the Movement

2.1 the Biplot

Consider commercial banks as an example. Different banks may hold differently structured financial asset portfolios and occupy different positions. We can assess a bank’s position based on the proportion of cash, commercial loans, mortgage loans, etc., as a percentage in the total asset (i.e., multiple dimensions). A biplot can provide us a quick intuition about how those dimensions relate to one another. However, the relative positions shown on biplot would not be accurate due to the loss of information in dimension reduction.

2.2 the 3-D and 2-D Scatterplots based on MDS

In an alternative approach, we can build distance-based features by exploring the inter-firm distance using multi-dimensional scaling (MDS). On Feb. 12th, 2018, Pacific Premier Bancorp launched an acquisition of Grandpoint Capital. Using this acquisition event as an example, we can first calculate the interfirm distance using MDS and then track how the acquisition shifts a bank’s position (in both a 3-D and 2-D scatter plots).

Dalio's Micro-economic 'Machine' and Short-term and Long-term Debt Cycles

Sat, 18 Apr 2020 00:00:00 +0000

1. Backgroud

Ray Dalio at Bridgewater Associates has provided a model (in narratives in his book and an animation video), which illustrates how the macro-economy and the banking system work in a mechanical way like a machine. I converted the model into a system map. Bridgewater’s performance and track record have proved the quality of his decision making, which allows him to be consistently correct in betting on what happens next in the economy.

2. Mapping the Micro-economic ‘Machine’

In this system, there are two essential components, the short-term debt cycle (the solid grey lines) and the long-term debt cycle (the red dotted lines). Debt creates money. Borrowing creates cycles. We enjoy economic growth until the system creates more credit than the real income-generating produtivity.

When interest rate hits zero, and the central bank can no longer stimulate the economy with a lower interest rate, we come to the end of the long-term debt cycle and enter into the stage of deleveraging and reflation (a.k.a. “the lost decade”). Next, we will wait and see if the government can execute a beautiful deleveraging (by controlling the rate of money-printing relative to the growth of debt).

the Positionings of Big Tech Firms on Industry Presence and Competition Description

Thu, 16 Apr 2020 00:00:00 +0000

In this post, I calculated the positioning of big tech firms based on two metrics, the multiple-industry presence (based on 3-digit SIC codes) versus the competition description (in 10-K filings), and compared the difference between the two approaches.

1 the positioning by industry presence across SIC categories

Scholars have used SIC codes to understand interfirm differences ¹ Big public firms typically straddle across multiple SIC industry categories. So it is straightforward and easy to see their positionings by multiple-industry participation. Accordingly, a firm’s position can be calculated as the sales distribution across multiple SIC industry categories ².

2 the positioning by competition description in 10-K filing

The SIC-code might be “backward-looking,” as firms tend to quite if particular combinations of multiple-industry business lines create economic inefficiencies. Only efficient multi-industry presence survives.

In contrast, the SEC filings may reflect managers’ current understanding of what they do. Especially when they characterize their competition, managers tend to draw their forward-looking expectations on the various forms of competitive threats. Some scholars³ and companies⁴ have used the text-based approach to calculate inter-industry distance.

2.1 calculating positioning based on the competition description

I collected the competition description texts of the largest 17 tech companies. Below are some frequently used words in those statements.

I ran an LDA topic modeling with 11 latent topics, which yields a low perplexity score.

The figure below shows how the 11 topics distribute across terms.

This is the topic loading on the 11 firms’ competition description, which is used to calculate the pairwise distance.

2.2 the final result: positioning by competition description

Here is the positioning map based on the competition description. The graph on the right shows the positioning of American companies. Looking at the results, we can see some interesting differences. The competition description moves some firms closer or further away from one another as compared to the SIC-presence positioning map.

for example:
Teece, D. J., Rumelt, R., Dosi, G., & Winter, S. (1994). Understanding corporate coherence: Theory and evidence. Journal of economic behavior & organization, 23(1), 1-30.

Lien, L. B., & Klein, P. G. (2013). Can the survivor principle survive diversification?. Organization Science, 24(5), 1478-1494.
↩
Litov, L. P., Moreton, P., & Zenger, T. R. (2012). Corporate strategy, analyst coverage, and the uniqueness paradox. Management Science, 58(10), 1797-1815.Lien, L. B., & Klein, P. G. (2013).↩
https://hobergphillips.tuck.dartmouth.edu ↩
Morningstart’s Company Comparables System ↩

Mapping Twitter Topics and Sentiment about FAAMG Companies since Covid-19 Outbreak

Wed, 15 Apr 2020 00:00:00 +0000

1 Motivation

On a whim, I mapped out what people say (the text) and feel (the sentiment) about the top tech companies (FAAMG) on twitter since the Covid-19 outbreak. The keywords and sentiment that are associated with the main topics should differ across these companies.

By no means I am drawing any causal association here. But the story starts with checking the recent price movements of the FAAMG stocks. Obviously, Covid-19 has impacted these companies quite differently.

2 Procedures

2.1 Data Source

For each company, I downloaded 1,000 tweets containing the company name as the hashtag (e.g. “#amazon” for Amazon). I used Python module “tweepy” to stream the data and set the “since” parameter to ‘2020-03-01’, which collects a sample of 1,000 tweets since March 1st.

2.2 Topic Modeling

Using Python module “gensim”, I detected three most significant topics (including Covid-19, business, and daily life), which are named by the associated keywords (most relevant terms).

Using Python module “vaderSentiment”, I computed sentiment scores for each tweet and aggregated the scores by company. The topic-based sentiment scores are used to construct the perceptual map, in which each company takes a different position. Using Python module “nltk”, I find the most frequent topic-based co-occurring words for each company.

3 Results

3.1 Perceptual Map of FAAMG Based on Topic Sentiment

It looks Amazon does well on all topics, especially on Covid-19-related topic. Microsoft and Apple do well on “Business”.

3.2 Keywords Co-occurrence Network on Each Topic

Here are the work keywords co-occurrence patterns for each topic of these companies.

Aim, Think, and Act: System Maps for Mental State

Tue, 31 Mar 2020 00:00:00 +0000

We all walk around and live a life based on a set of internal assumptions and logics in our minds, which continuously process the information from the external reality and produce the results in life. If you don’t like the results in your life, simply change your mind, and life will change for you.

The three system-maps below illustrate three types of self-motivators (about future, past, and present) and the corresponding pitfalls to avoid. Remember, if you don’t take effort to make your mind work in a better way, it will continue to work in the old way, keep falling into the pitfalls, and produce the same results that you hate.

Aim High and Prevent ‘Stopping Search’

Keep aiming high.
Search more (frequency) and search farther (space).
Don’t estimate the time needed for making improvement based on the current solution.

The first type of self-motivator is goal-setting. For the future, not only you need to have a CLEAR goal, but also you need to make it CHALLENGING¹ and even seem impossible². Aiming high generates aspiration, which will trigger SEARCH behavior³. The more you keep aiming high, the more frequently and farther you search for alternative solutions. It is natural and typical that you don’t find a solution each time you search. But the more you search, the more likely you will experience an epiphany and find the ONE idea among a series of many ideas to improve your condition effectively and drastically. Keep aiming high, because the fulfillment of your aspiration from an improvement of your condition will also weaken your aspiration.

The pitfall: not to repeatedly and frequently energize your aspiration by keeping a clear and challenging goal, because it may seem to take a long time. You develop a false belief when you estimate the time needed to make improvement based on the current rate of progress. It seems to take a long time because you have not yet found the solution to expedite your growth. Finding a solution requires more search. More search is driven by your aspiration, which is energized by your goal. Take as much time as you need to set your goal. Write it on your wall.

Think Deep and Prevent ‘Shifting Burden’

Employ fundamental solutions.
Expect the DELAY to use the right solution to solve the right problem.
Avoid quick fix (e.g. instant gratification).

The second type of self-motivator is problem-solving. Your success is the way you spend your time doing your BEST—the way you take the talent you were born with and all the knowledge and skills you’ve since developed in the past and using them toward a problem at present. We have limited mental capacity to mobilize rationality and need time to reason⁴. It takes time for us to think through to find the most important problem to solve. It also takes time to find the fundamental solution—the right one. If you use the right solution to solve the right problem, you also solve all the problems. It is once and for all.

The pitfall: not to take time and think deep enough to find the fundamental solution. This critical step only requires patience. But since there is a DELAY for the fundamental solution to takes effect in addressing the problem, your emotional thinking and physiological drive will offer a quick fix. The less patience you have, the more you are tempted to use the quick fix. However, the quick fix, such as an instant gratification or a duct tape, can only solve your current problem by shifting the burden to the near future—leaving you with more problems to solve. Think hard before you act.

Act Fast and Prevent ‘Eroding Goals’

FOCUS on the action.
Expect the DELAY.
FIX the goal (and never compromise).

The third type of self-motivator is action-taking. A quote by Jack Welch: “you pick a general direction and implement it like hell.” Expect the DELAY for the action taken to take effect. Live in the PRESENT. Focus on the task flow⁵, in which you are fully immersed in a feeling of energized focus and enjoyment in the process. This is a simple 3-step process: fix your goal in your mind, rest your focus in the action, and let the action take its course to improve the condition.

The pitfall: not to focus on the action with a fixed goal but too much on the gap between your current condition and your goal. It is easy to think fast than act fast. Usually when people don’t act fast, they think too much. The more you move your mental energy away from the action to worry about the condition, the more psychological stress you imprudently develop, urging you and tempting you to compromise your goal. When you tune your mind into such a compromise channel, the delay for action to improve condition will seem forever and become so hard to endure. If you don’t like a mixed feeling of relaxing and helpless, fix your goal and focus on your action, not your condition.

Locke, E. A., & Latham, G. P. 2002. Building a practically useful theory of goal setting and task motivation: A 35-year odyssey. American Psychologist, 57(9): 705–717.↩
Collins, J. C., & Porras, J. I. 1996. Building your company’s vision. Harvard Business Review, 74(5): 65.↩
Cyert, R. M., & March, J. G. 1963. A behavioral theory of the firm. Englewood Cliffs, NJ, 2.↩
Simon, H. A. 1997. Models of bounded rationality: Empirically grounded economic reason, vol. 3. MIT press.↩
Nakamura, J., & Csikszentmihalyi, M. 2014. The concept of flow. Flow and the foundations of positive psychology: 239–263. Springer.↩

An Ongoing Collection of Strategy Matrices

Tue, 31 Mar 2020 00:00:00 +0000

Here is a collection of analytical matrices in strategic management and entrepreneurship. As a tool for strategic decision making, each matrix is made of two analytical dimensions.

1 General-purpose Matrices

SWOT Analysis [analytical dimensions: external/internal × harmful/helpful]
TOWS Analysis [analytical dimensions: (strength vs. weakness) × (opportunities vs. threats)]
Eisenhower Priority Matrix [analytical dimensions: urgency × importance]
Risk Reward Matrix [analytical dimensions: risk × reward]
Skill-Will Matrix for Managing & Leading People [analytical dimensions: people’s will × people’s skill]

2 Strategic Management

2.1 Business Strategy

Generic Strategies [analytical dimensions: market scope × profit driver]
Miles & Snow’s Strategic Type [analytical dimensions: environment × action]
BCG Matrix [analytical dimensions: market growth × market share]
Blue Ocean Strategy: Four-Action Matrix [analytical dimensions: relative performance × perceived value]
Competitor Identification Matrix [analytical dimensions: market overlap × resource similarity]

2.2 Corporate Strategy

Ansoff Growth Matrix [analytical dimensions: product (old vs. new) × market (old vs. new)]
M&A Integration Matrix [analytical dimensions: interdependence (acquirer v. target) × autonomy (target)]
Outsourcing Matrix [analytical dimensions: strategic impact × cost benefit]
Kraljic Purchase Control Matrix [analytical dimensions: strategic impact × cost benefit]
Bartlett & Ghoshal Globalization Matrix [analytical dimensions: global standardization (↓cost) × local customization (↑value)]
Stakeholder Interest Influence Matrix [analytical dimensions: stakeholder power × stakeholder interest]

3 Innovation Strategy

Types of Innovation [analytical dimensions: problem maturity × solution maturity]
Perrow’s Technology Typology Matrix [analytical dimensions: task analyzability × task variability]
Perrow’s Coupling-Interaction Diagram [analytical dimensions: system coupling (tight vs. loose) × system interaction (linear vs. complex)]
Product-Process Change Matrix [analytical dimensions: product change × process change]
Abernathy & Clark Innovation Matrix [analytical dimensions: market creation (low vs. high) × technological change (low vs. high)]
How-Wow-Now Matrix [analytical dimensions: novelty × feasibility]
Verganti’s Innovation Matrix [analytical dimensions: new technology × new meaning]
Standford Human-Systems Design-thinking Matrix [analytical dimensions: creation (understand vs. create) × abstraction (concrete vs. abstract)]

Didn’t See the Matrix You Use? TELL ME to add it!

Costoco

Fri, 03 Jan 2020 00:00:00 +0000

Costco’s membership renewal rates in the united states and Canada have remained at roughly 90%. The traffic-driving value that Costco offers in its stores is fueled by cost leverage that, in turn, feeds additional store visits.

Costco has achieved a competitive advantage based on its intangible assets and cost leadership. The firm’s membership subscription model has held steady through the financial crisis and the rising competitions, especially Amazon’s Prime memberships. In addition to cost management and distribution leverage, Costco’s low prices are enabled by its procurement strength.

Costco sells roughly 3,700 different stock-keeping units (SKUs) in its stores, a fraction of the 75,000 to 80,000 at Walmart in the United States, which offers a significantly greater number of items via its marketplace. Costco’s sales per SKU were over 30 million, towering above Walmart’s less than 3 million and Target’s $1 million, contributing to its procurement cost advantage.

Costco benefits from a number of traffic drivers, particularly its food offerings (53 % of fiscal 2019 sales) and fuel. Fuel sales serve as a meaningful counterpoint to traffic pressure from digital sellers. With in-store prepared food offerings and a well-developed private-label portfolio, Costco’s store offering is well distinguished. With in-store ready food offerings, the Kirkland signature brand accounts for around a quarter of Costco’s sales and is also margin-accretive.

Costco has been increasing its e-commerce efforts (roughly 4% of sales), focusing on items not available in stores (including travel and other services) and a distinguished product novelty. The digitization of retail has strained traditional sellers. The pressure also intensifies on traditional competitors to provide omnichannel solutions.

Some risk factors: (1) Costco has thrived under a variety of conditions, but as competition increases and it expands into new markets, the degree of difficulty also rises. (2) Costco handles a significant amount of customer data, and a breach could expose it to financial and reputational risk. (3) Several costs that are outside Costco’s control include wages (particularly its gaps against competitors), currency, trade policy, and the broader macroeconomic environment.

3-D Printing

Fri, 29 Nov 2019 00:00:00 +0000

Overview of the ‘3-D Printing’ System

Overview of the ‘3-D Printing’ System

Additive manufacturing (3-D Printing) beats subtractive manufacturing (traditional manufacturing) by reducing cost and increasing flexibility at the same time. If the natural selection favors the more efficient production mode that generates less economic waste, 3-D printing will be the sure winner.¹ In essence, 3-D printers can mass-produce customized physical objects. This process is also called “mass customization,” which breaks the cost-value trade-off and shifts the productivity frontier outwards.
Traditionally, assembly lines are intended to utilize standardized modules to construct complex systems². However, the modularity in subtractive manufacturing is so imperfect that the supply chain carries the organizational costs from the modulization (intra-firm coordination) to the assembling (inter-firm transaction) processes³. As a result, the traditional supply chain is comprised of specialized firms handing standardized modules⁴. Since such specialized firms can not be efficient without a stable external demand to match standardized internal processes, the entire supply chain will be too rigid to meet the changing demand or transfer the “rigidity” cost to specialized firms.
Mass customization breaks the standardization-customization compromise. The technology of 3-D printing reverse-engineers the problem of supply chain rigidity. Instead of preparing modules, manufacturers can act directly on demand and shorten the cycle between the point of consumption and point of production⁵. The middlemen are eliminated as the module-producing and assembling processes are replaced by a single data input file for a 3-D printer. The end product can be both scalable and personalized, designed to meet the demand.

Winter, Sidney G. Economic" natural selection" and the theory of the firm. Vol. 4. Institute of Public Policy Studies, University of Michigan, 1964.↩
Simon, Herbert A. “The architecture of complexity.” Facets of systems science. Springer, Boston, MA, 1991. 457-476.↩
Newman, Stephen T., et al. “Process planning for additive and subtractive manufacturing technologies.” CIRP Annals 64.1 (2015): 467-470↩
Williamson, O. E. “1975: Markets and hierarchies. New York: Free Press.” (1975).↩
Mertz, Leslie. “Dream it, design it, print it in 3-D: what can 3-D printing do for you?.” IEEE pulse 4.6 (2013): 15-21.↩

Block Chain

Fri, 22 Nov 2019 00:00:00 +0000

Overview of the ‘Block Chain’ System

Overview of the ‘Block Chain’ System

If natural selection optimizes for efficiency ¹ and entropy in the universe is ever-increasing², there will be a trade-off between the order of a social structure and the freedom of its interacting agents. A social system constrains individual freedom according to the limitation of agentic inter-connectivity ³. For example, when we rely on market transactions to exchange (integrate) specialized productions (divided labor), we give power to intermediaries (firms, government, communities) to fill in the gap in inter-connectivity (in which intermediaries connect with unconnected participants). However, when everyone is connected with everyone else, we don’t need to rely on intermediaries to maintain order.
The full connectivity creates a new social structure. As social structure serves as the medium for agency, the new social structure will enable us to achieve what’s impossible in the past ⁴. Because all participants are interconnected with one another, the entire system prevents a central intermediary from manipulating the flow of information. As there is no information asymmetry, ownership contracts will be secured (automatically verified) and transparent ⁵.
Free from information brokerage and the need for trusted third-parties to secure transactions, the ownership contracts associated with any virtual usage or value can facilitate transactive collaborations among autonomous agents. The costs (and needs) for trust or order are eliminated at the system level across the fully connected peer-to-peer network. The efficiency gain created by specialization (by division of labor and separation of tasks) is no longer bounded by the extent of market exchange for integrating unconnected agents. System integration will directly derive from the complete inter-connectivity to enhance system-wide wealth.

Hannan, Michael T., and John Freeman. “The population ecology of organizations.” American journal of sociology 82.5 (1977): 929-964.↩
Denbigh, Kenneth George, and Kenneth George Denbigh. The principles of chemical equilibrium: with applications in chemistry and chemical engineering. Cambridge University Press, 1981.↩
Simon, Herbert A. “The architecture of complexity.” Facets of systems science. Springer, Boston, MA, 1991. 457-476.↩
Anthony Giddens. The constitution of society: Outline of the theory of structuration. Univ of California Press, 1984.↩
Nakamoto, Satoshi. Bitcoin: A peer-to-peer electronic cash system. Manubot, 2019.↩

Digital Marketing

Fri, 15 Nov 2019 00:00:00 +0000

Overview of the ‘Digital Marketing’ System

Overview of the ‘Digital Marketing’ System

Digital marketing utilizes new forms of media to increase the content ROI relative to the traditional media (e.g., commercials on mass media). In the traditional mode, it is not only costly to create rich content catering to different tastes but also impossible to control the traffic-conversion rate in the long tails ¹. Digitization reduces content creation cost and enables niche content creators to expose their content to the entire global audience. Niche content is more effective and efficient in traffic generation and lead conversion because of greater relevancy and closer relationship with niche audience.
A key component of the digital marketing process is user data. User data live on the consumer side of the content exchange process. Traditional content relying on one-sided communication fails to connect with the differing consumer demands, preferences, and needs ². By contrast, in each vertical niche, connections and relationships can be repeated and interactive. Continuous user feedback becomes readily available from the recursive exchange of real, unmet human needs. During the exchange process, valuable user data reveals profitable problems and new solutions, which is either unavailable or costly to collect through the traditional channel.
The instant feedback also comes directly from the ROI of the content creation and user interaction process. Digital marketers can directly read the metrics on a dashboard and relate such metrics to their content’s marketing performance ³. The learning cycle will dramatically decrease for digital marketers to quickly pivot their content or find better product matches to improve content performance. In digital marketing, quick learners can scale their content creation and monetization system more quickly.

Anderson, Chris. The long tail: Why the future of business is selling less of more. Hachette Books, 2006.↩
Kamins, Michael A., et al. “Two-sided versus one-sided celebrity endorsements: The impact on advertising effectiveness and credibility.” Journal of advertising 18.2 (1989): 4-10.↩
Morris, Neil. “Understanding digital marketing: marketing strategies for engaging the digital generation.” (2009): 384-387.↩

E-commerce

Fri, 08 Nov 2019 00:00:00 +0000

Overview of the ‘E-commerce’ System

Overview of the ‘E-commerce’ System

E-commerce allows the transactions of goods and services to be conducted electronically on the internet. The ease of electronic transaction transforms the organization form due to the reduction in transaction cost ¹. For consumers, online shopping reduces the search cost for desired products. For sellers, the reduced communication cost enables businesses to outsource core functions (such as marketing, inventory, distribution, etc.) to e-commerce platforms or external providers.
The supply chain is deconstructed to allows e-commerce participants to easily switch transacting partners both up-stream and down-stream. Core business functions become easily accessible “commodities,” offered by specialized platforms (marketing cloud, on-demand inventory, delivery services) ². The efficiency boost from business-processes outsourcing and automation enables sellers to focus more on selling and responding quickly to market demand. Meanwhile, platforms replace traditional supply chains, enjoying significant network effects and high customer switching costs.
Because it is no longer efficient to perform many business functions in-house, value chains decompose as businesses reorganize to compete primarily on their effectiveness in selling. This gives power to the consumers facing an increasing number of options to choose from. However, increasing options also bring about information overload, in various vertical categories. Consequently, content influencers reintermediate the selling process with richer content to get closer to the audience in a particular market segment. However, content richness in one segment reduces the reach for many, driving the ever-increasing content fragmentation ³.

Williamson, O. E. “1975: Markets and hierarchies. New York: Free Press.” (1975).↩
Markus, M. Lynne, and Claudia Loebbecke. “Commoditized digital processes and business community platforms: New opportunities and challenges for digital business strategies.” Mis Quarterly 37.2 (2013): 649-653.↩
Wurster, Thomas S. Blown to bits: How the new economics of information transforms strategy. Harvard Business School Press, 1999.↩

Internet of Things

Fri, 01 Nov 2019 00:00:00 +0000

Overview of the ‘Internet of Things’ System

Overview of the ‘Internet of Things’ System

Specialization creates efficiency. Integration increases wealth. In modern societies, specialized “things” — physical objects created by specialized producers — result in the lack of integration among the divided things (and thus leave money on the table). The total wealth in an economy lies in the efficiency boost from specialized producers, which ultimately need to bring their products together through market exchange mechanism. ¹
New combinations generate new innovation. Innovation resides in the new linkages or new combinations between exiting things ². In the digital age, new connections between existing things create new data dimensions, which leads to new cost-saving and value creation opportunities.
Right after new connections among existing things generate stabilized benefits for existing users and producers, the newly integrated system starts to form new sub-systems at higher-level complexities ³. Again, new cost-saving and value-creation opportunities will emerge from the higher-level connections among newly formed sub-systems.

Smith, Adam. The Wealth of Nations: An inquiry into the nature and causes of the Wealth of Nations. Harriman House Limited, 2010.↩
Schumpeter, Joseph A. Capitalism, socialism and democracy. routledge, 2013.↩
Simon, Herbert A. “The architecture of complexity.” Facets of systems science. Springer, Boston, MA, 1991. 457-476.↩

Platform

Wed, 30 Oct 2019 00:00:00 +0000

Overview of the ‘Platform’ System

Overview of the ‘Platform’ System

Platforms play the “game of power” by leveraging network effects. Power can be created through a virtuous cycle in which increased numbers of platform users continuously increase user experience and disproportional user dependency on the platform ¹. As long as a platform can dominate a vertical field, the network effects will ultimately pay off by reducing the marginal cost for user acquisition and user retention to zero, and increasing switching cost on the user side to infinity. We often call this phenomenon “winner take all” ². As a result, platforms subsidize the users or the third-party sellers (complementors) until the asymmetric dependency is established.
Single-sided network effect (or same-side network effect) occurs when a focal user can interact with a greater number of others joining the same network and using the same platform. Telephone, internet, and online social networks (almost any technology) are all this kind. As Metcalfe’s law states, the possible interconnections within the network are proportional to the square of the total number of connected users ³. After widespread adoption, users will consider the platform as an indispensable commodity. The platform enjoys significant economy of scale.
Multi-sided network effect (or cross-side network effect) occurs when a focal user can derive greater value when a greater number of complementors joining the platform ⁴. At the system level, Multi-sided network effects reduce the total transaction costs in the multi-sides markets by reorganizing the functional components within a traditional value chain. ⁵ By reconstructing social and economic networks, platforms can deconstruct a value chain into separate parts, attract participants (and pool resources) from each part, and utilize multi-sided network effect to leverage the scale and asymmetric dependencies over individual users/complementors.

Casciaro, Tiziana, and Mikolaj Jan Piskorski. “Power imbalance, mutual dependence, and constraint absorption: A closer look at resource dependence theory.” Administrative science quarterly 50.2 (2005): 167-199.↩
Eisenmann, Thomas, Geoffrey Parker, and Marshall W. Van Alstyne. “Strategies for two-sided markets.” Harvard business review 84.10 (2006): 92.↩
Metcalfe, Bob. “Metcalfe’s law after 40 years of ethernet.” Computer 46.12 (2013): 26-31.↩
Parker, Geoffrey G., and Marshall W. Van Alstyne. “Two-sided network effects: A theory of information product design.” Management science 51.10 (2005): 1494-1504.↩
Rochet, Jean‐Charles, and Jean Tirole. “Two‐sided markets: a progress report.” The RAND journal of economics 37.3 (2006): 645-667.↩

Open Source

Fri, 25 Oct 2019 00:00:00 +0000

Overview of the ‘Open Source’ System

Overview of the ‘Open Source’ System

Organizations are often instruments of purpose. ¹ Organizations create efficiency by getting individuals to work towards a common goal. However, organizations may confine productivity and block free knowledge flows when a purpose is not commonly shared and clearly defined. ²
Emerging new technologies tend to benefit from free knowledge flows across organizational boundaries. In the case of innovation, organizational boundaries become the barriers for good ideas to connect and interact. Thus, proprietory ownership of good ideas not only prevents external minds from connecting with internal minds, but also causes internal minds to miss out on the opportunity of good ideas from external inter-connected minds. Opportunity cost can quickly override the ownership benefits and render the good in-house ideas obsolete.
Without the constraint of organizational boundaries, good ideas can freely inter-connect across organizations, locations, and industries to form new ones, which are otherwise impossible if the source code is “closed” for narrow and pre-defined commercial interests. ³ With free connections between new ideas and the self-regulation protocols within open-source communities, new technologies and new use cases constantly emerge. Thus, open-source developers can continuously crowdsource good ideas and conveniently build on modularized packages to create higher-level complexities.

March, James G., and Robert I. Sutton. “Crossroads—organizational performance as a dependent variable.” Organization science 8.6 (1997): 698-706.↩
Valentine, Melissa A., et al. “Flash organizations: Crowdsourcing complex work by structuring crowds as organizations.” Proceedings of the 2017 CHI conference on human factors in computing systems. 2017.↩
Lerner, Josh, and Jean Tirole. “Some simple economics of open source.” The journal of industrial economics 50.2 (2002): 197-234.↩

Robotics

Fri, 18 Oct 2019 00:00:00 +0000

Overview of the ‘Robotics’ System

Overview of the ‘Robotics’ System

Because robotics are built with standardized programs, robotics can leverage the scale curve. The marginal production cost reduces as the volume increases. Yet, with integration with A.I., robotics are now being programmed to perform more and more complex tasks. Enhanced programmability will generate more and more value creation at a constant production cost. ¹
Compatible with A.I. ², robotics can fulfill highly customized needs. This means A.I. powered robotics can learn from data and become smarter over time. With better user experience, A.I.-powered robots have the opportunity to spend more time interacting with human users, which creates opportunities to extract user data for improving the performance in fulfilling personalized needs.
Data integration across different users or across different use cases will accelerate the learning curve of robotics to a higher rate. Robot-to-robot connections allow inter-robotic collaborations in creating enhanced solutions. ³ As a result of increasing adoption, user-robot interactions will continuously generate user feedback, which, in turn, continuously reveal unsolved problems and create new business opportunities.

McKenna, Regis. “Marketing is everything.” (1991): 65-79.↩
Kober, Jens, J. Andrew Bagnell, and Jan Peters. “Reinforcement learning in robotics: A survey.” The International Journal of Robotics Research 32.11 (2013): 1238-1274.↩
Mondada, Francesco, et al.“The cooperation of swarm-bots: Physical interactions in collective robotics.” IEEE Robotics & Automation Magazine 12.2 (2005): 21-28.↩

Social Media

Thu, 17 Oct 2019 00:00:00 +0000

Overview of the ‘Social Media’ System

Pescosolido, Bernice A. “Beyond rational choice: The social dynamics of how people seek help.” American journal of sociology 97.4 (1992): 1096-1138.↩
Podolny, Joel M. “Networks as the pipes and prisms of the market.” American journal of sociology 107.1 (2001): 33-60.↩
Katz, Michael L., and Carl Shapiro. “Systems competition and network effects.” Journal of economic perspectives 8.2 (1994): 93-115.↩

Uber

Fri, 11 Oct 2019 00:00:00 +0000

Uber ’s position in the autonomous vehicle race could equalize gross and net revenue, after no longer needing to pay drivers. Uber technologies have matched riders with drivers completing trips over billions of miles. At the end of 2019, Uber had 111 million users who used the firm’s ridesharing or food delivery services at least once a month.

Uber helps people get from point A to point B by taking ride requests and matching them with drivers available in the area. The firm refers to this as personal mobility, which also shorter - distance transportation via electronic bikes and scooters. Uber faces stiff competition from players such as Lyft (in the U.S.) and didi, a business in which uber has a 20 % holding after the sale of its operations in China to didi in 2016. while uber no longer operates in China, it does compete with didi in other regions around the world. The market remains fragmented, and uber competes with many local ridesharing platforms and taxis. Globally, the market is fragmented. Uber has a 30% global market share and will be the leader in the total addressable ridesharing market ( excluding China ) by 2024.

The firm ’s food delivery service will be one of the main revenue growth drivers for the firm as it will benefit from cross-selling to its large ridesharing user base. Further, utilize of uber ’s overall on-demand platform can also help the firm progress toward profitability. Uber has been in talks to acquire Grubhub in an all-stock deal for each Grubhub share. The addition of GrubHub could strengthen the supply and demand sides of uber ’s network effect moat source. On the supply side, the firm would be better able to attract and retain restaurants. With more food delivery requests, uber is likely to also maintain more of its ride-hailing drivers. Also, this may not only reduce diner acquisition costs for uber but will also allow uber to more effectively cross-sell its two main businesses to a larger user base.

The ridesharing platform benefits from network effects and valuable intangible assets in the form of user data. These sustainable competitive advantages will help uber to become profitable and generate excess returns on invested capital. Uber ’s network effects benefit drivers and riders, creating a continuous virtuous cycle. As the number of drivers increased, the timeliness and reliability of the service improved. The riders on the platform benefit as more drivers are added, and existing drivers benefit from more riders, making the driver ’s use even more efficient. The increase in vehicle capacity usage is growing in the average number of rides dispatched per unique uber vehicle.

As uber benefits from its network effect, it gains access to valuable intangible assets in the form of user data, which helps the firm improve its services. In turn, Uber ‘s service may become more effective as it further monetizes real-time supply and demand-driven pricing. Uber may also use extensive data and knowledge to tap into other markets. Uber may use this data to tap into other markets. The firm may also increase its vehicles’ capacity usage. The firm compiles data from the rider app about the locations users request rides to and from and at what times of the day. When combined with the user-generated driver ratings, we think such information helps uber improve the timeliness of matching riders with drivers. Such overall enhancement in service could help the firm strengthen its network effect by increasing users and ride requests per user, which helps uber gather additional data.

Virtualization

Tue, 08 Oct 2019 00:00:00 +0000

Overview of the ‘Virtualization’ System

Overview of the ‘Virtualization’ System

Consider a form of commodity that is under private ownership but does not enter the market exchange, like the electricity coming from a private generator. Because electrical energy would be the same from different generators, electricity can be pooled together to create greater efficiencies than private generators. Thus, in most societies, individual households are renting electricity from public utility companies instead of owning it. In this case, utility companies create a virtual version of electricity resources for households, which do not own real generators. Similarly, there is room to improve societal efficiency by converting some forms of ownership to rentorship, especially when private ownership creates underutilized resources.
When individual companies perform similar business functions using the computer, it is possible to improve societal efficiency by pooling the computing and storage resources together and create a virtual instance of a greater computer system. For individual companies, renting the virtual instances to have applications running on top of it is more efficient than owning (and maintaining) the actual hardware. Also, virtual instances create the flexibility for different companies with customized needs to experiment with new applications, without allocating a budget to acquire hardware before the profit potential of new applications becomes predictable.
When A.I. becomes the new electricity, cloud computing becomes the new power station, pooling resources creates on-demand availability for businesses and individuals ¹. Similar to the resource pooling logic, but in a different efficiency-improving mechanism, under-utilized assets can be pooled together from (and shared by) the supply side to better match with demand ². In this case, renting is still more economical than owning, but the societal cost reduction comes from the information brokerage.

Kumar, Sanjay, et al. “vManage: loosely coupled platform and virtualization management in data centers.” Proceedings of the 6th international conference on Autonomic computing. 2009.↩
Schor, Juliet. “Debating the sharing economy.” Journal of Self-Governance and Management Economics 4.3 (2016): 7-22.↩

Salesforce

Fri, 27 Sep 2019 00:00:00 +0000

Salesforce dominates the salesforce automation (SFA) and customer relationship management (CRM) space and controls 30% in a highly fragmented market by the end of 2019. The company has added legs to the business clouds, including customer service, marketing automation, e-commerce, analytics, and artificial intelligence. The management has put more emphasis on expanding margins across the clouds.

Salesforce has assembled a front-office empire that it can build on for years to come. Thanks to the company’s revolution, business software is increasingly accessed through a web browser and delivered over the internet, so inventing the Software-as-a-service (SaaS) software delivery model. Salesforce also offers customers a platform-as-a-service (PaaS) solution, complete with the AppExchange, to rapidly create and distribute apps. This further strengthens the substantial community of Salesforce users.

Salesforce is widely considered a leader in its served markets. The tight integration among the solutions and the natural fit they have with one another makes for a powerful value proposition. Indeed, more than half of enterprise customers use multiple clouds. Moreover, customer retention has gradually improved over time.

Salesforce paved the way for the software industry as it exists now by first selling the concept of SaaS to prospective customers and then selling the actual SFA and CRM products. The SaaS model, and SFA and CRM applications, rose in popularity as customers were able to avoid high up-front costs. SaaS vendors benefited from predictable revenue streams, the elimination of piracy, and supporting only one product version.

Salesforce has gone from no product to 33% market share over the last 20 years. Customers view salesforce as the clear front-runner in a category that increases the productivity of sales representatives. The mission-critical software indeed helps drive revenue for users. Also, customers are reluctant to switch away from the sales cloud because of the time, expense, and lost productivity of retraining the workforce on a new platform. The corporate risk of making a change is high, as executives engage in self- serving behavior. In fact, executives can jeopardize their own careers by pushing to switch from a leading solution that is functioning well and meeting their corporate needs.

Other than the sales cloud, service cloud includes a set of solutions aimed at helping an enterprise deliver customer service and support at scale. Customer service is another mission-critical function that directly helps a company retain customers. Similar to the sales cloud, customers are also reluctant to invest the time and expense of converting a critical revenue-driving function from one application to another. It is often more expensive to find a new customer than it is to retain an existing customer.

Salesforce has initiated the marketing cloud as one segment under is business clouds umbrella. Marketing automation creates mass-customized cross-platform campaigns to targeted audiences. The company was early to the market with a cloud-based application development platform for customers. Use of the platform provided the low-investment benefits of a SaaS product, allowing for immediate and smooth integration with salesforce.com’s solutions, in which developers can sell or give away the applications they developed on the AppExchange. A variety of publicly traded companies started out as apps developed on the platform and initially distributed on the AppExchange. In fact, AppExchange was a novel idea and predated Apple’s App Store by two years as it was revolutionary and enticing to developers.

Netflix

Fri, 20 Sep 2019 00:00:00 +0000

Netflix has continued to burn billions of dollars of cash to create its original content with no end in sight. The need for increased content and marketing spend outside of the U.S. will limit the rate of margin expansion for the international segment. However, the volume of sales of the company has continued to grow over the past years.

Netflix has used its scale to construct a massive data set that tracks every customer interaction. The firm then leverages this customer data to better purchase content as well as finance and produce original material. Netflix has expanded rapidly into markets abroad as the service has more subscribers outside of the U.S.

Media firms will continue to reap the benefits of both an additional window for existing content and another platform for new content. Larger firms like Disney and WarnerMedia have or will soon launch their own SVOD platforms to compete against Netflix. Such a usage pattern may constrain Netflix’s ability to raise prices without inducing greater churn. Netflix has also expanded further into local-language programming to offset the weakness of its skinny offering in many countries.

New content not only strengthens the relationship with current customers but also attracts new customers via word of mouth and the halo effect from critical acclaim and award nominations. The rapidly growing subscriber base (over 130 million worldwide) creates a humongous data set that Netflix mines to better purchase and create content. Netflix tracks every customer interaction, from large (total time spent at Netflix) to minute (whether a user pressed fast forward). By the end of 2019, Netflix accounted for 26% of all global video streaming traffic, beating out YouTube at 21% and amazon at 6%. The average Netflix user worldwide watches more than 90 minutes of video per day. Netflix can search for this information to better understand the network and device performance, customer behavior, and content popularity.

Netflix analyzes data traffic, video performance, and buffering to better understand where data loss and slowdown occurs. The company also examines specific subscriber actions by type of action and device used to formulate a better user interface and to tweak device-specific applications. The real-time nature of the data provides Netflix with the ability to iterate more quickly than traditional user groups or beta testing methods. A large number of subscribers using different devices across multiple countries generate an extensive, growing, robust data set.

The content discovery engine provides recommendations based on a subscriber’s previous viewing habits in context with similar viewing habits. While growing rapidly as a streaming video provider, the company understood the need to create original content to distinguish its offering. An often-cited example of this data is “House of Cards,” an adaption of a British miniseries starring Kevin Spacey. Netflix noted that the director David Flincher’s movies were generally watched from beginning to end, that Spacey’s films had performed well, and that the original version was popular with subscribers.

As Netflix is now available in almost every country, this cost could explode, particularly as new competitors like Disney enter the market and ramp up their content spending. The cost of licensing content will also rise as competitors emerge and bid for content that Netflix desires. The move to more original content adds costs and risks. Netflix ’s expansion outside the U.S. could continue to drag on cash flow due to different tastes and lower video consumption. The cost to deliver content could increase, and the need to pay for fast-lane network access could drag on margins. However, increasing price rates could limit growth and increase subscriber churn.

Microsoft

Fri, 13 Sep 2019 00:00:00 +0000

The shift to subscriptions accelerates growth after the initial growth pressure. The centerpiece of the new Microsoft grew at a staggering 92 % rate in fiscal 2018. It offers customers a painless way to experiment and move select workloads to the cloud. Microsoft can also leverage its massive installed base of all Microsoft solutions as a touchpoint for a strategic move. Microsoft continues to launch new services centered around these broad themes, as it continues to launch new services centers around these broad themes. Microsoft is an excellent launching point for secular trends in A.I., business intelligence, and the internet of things.

Office 365 (the cloud-based version of the traditional perpetual license Microsoft office) retained in office productivity software, which will remain steady in the foreseeable future. Customers will continue to drive the transition from on-premises to cloud solutions, and revenue growth will remain robust. The initial move to the cloud was painful, as both revenue and margins dropped. However, Microsoft’s revenue has accelerated, thanks to the cloud transition.

An office productivity suite generally consists of spreadsheet and presentation software applications. Microsoft offers a variety of versions of the office 365 suite and increasingly fewer perpetual license versions. Office 365 starts at approximately $6 per month and tops out at roughly $35 per user per month. The perpetual license version is $150. Office 365 already has more than 165 million subscribers.

Microsoft has continued to enjoy a dominant market share position in office suites, with Google being the only other vendor of consequence. Many users are willing to pay a minimum of $70 per year to use office 365 when free versions that are generally similar in terms of features and interface are available. Microsoft office benefits from high switching costs. Because of the significant installed base, it would be highly disruptive for a company to pivot to an office suite other than office 365. For example, reports within the financial function of countless enterprise users are pulled from an oracle, or other popular databases into an excel file that can then be manipulated.

Microsoft Dynamics is a cloud-based enterprise resource planning (ERP) and customer relationship management (CRM) suite of applications designed to help mid-sized companies, or divisions of larger companies, run their businesses. Dynamics accounts for approximately 2% of total revenue and is growing in the low double-digit area. The Dynamics revenue base is shifting from a perpetual license and maintenance model into a subscription model. As such, Dynamics has increased its profile and is slowly gaining share market share.

Microsoft’s presence in the P.C. market with both its O.S. and office productivity software allowed it to easily enter the server market. Today the I.T. backbone of many of the largest companies in the world is built on Microsoft server. Hence, replacing any part of the core of an enterprise’s I.T. environment would be a significant undertaking for any company in terms of cost, time, and risk. The early lead and substantial market share led to a wide variety of developers joining the ecosystem bringing in applications, middleware, and development tools.

In an IaaS model, the provider offers the hardware virtualization, networking, and storage as a computing service delivered over the internet. As software is added, IaaS quickly becomes PaaS. The provider also offers and hosts operating systems, middleware, and core I.T. applications. Currently, Microsoft Azure, Amazon AWS, and Google Compute are leaders in this category.

Microsoft has used its presence to attract more users to its user base, which in turn attracts more developers in a virtuous circle. Developers have a lot invested in learning certain languages and how to effectively write software under a given umbrella. It would be a time-consuming, and therefore costly endeavor to learn additional languages on a different platform. The GitHub acquisition (closed on October 26, 2018) bolsters Microsoft’s position for developers. Google, Apple, Amazon, and Microsoft each have a material presence on GitHub and use it for documentation of code. It remains to be seen if these mega-cap tech competitors will allow their developers to continue to use GitHub.

Nvidia

Fri, 06 Sep 2019 00:00:00 +0000

The increasing complexity of graphics processors provides a barrier to entry for most potential rivals, as it would be challenging to match Nvidia’s large R&D budget. The firm has a first-mover advantage in the autonomous driving market that could lead to widespread adoption of its self-driving platform. The ongoing expansion of artificial intelligence and deep learning that rely on Nvidia ’s graphics chips presented the firm with a potentially massive growth opportunity.

PC gaming enthusiasts generally purchase high-end discrete GPUs offered by the likes of Nvidia and AMD. This involves large swaths of data followed by techniques that develop algorithms to produce conclusions in the same way as humans. Today’s basic variants of AI are consumer-based digital assistants, image recognition, and natural language processing.

Nvidia’s DRIVE PX platform is a deep learning tool for autonomous driving. It is being used in research and development at more than 370 partners. The company views the car as a supercomputer on wheels, although this segment currently contributes relatively little to the top line. Nvidia has patents related to the hardware design of its GPUs in addition to the software and frameworks used to take advantage of GPUs in gaming, design, visualization, and other graphics-intensive applications. The latest pc games typically require system software updates that optimize the performance of GPUs.

Nvidia has a cost-leadership advantage and intangible assets related to the design of graphics processing units (GPUs). The firm is the originator of and leader in discrete graphics, having captured the lion’s share of the market from longtime rival AMD. The market has significant barriers to entry in the form of advanced intellectual property, as even chip leader intel was unable to develop its own GPUs despite its vast resources and ultimately needed to license IP from Nvidia to integrate GPUs into its PC chipsets. Nvidia has gained share at the expense of AMD as gamers have moved from mainstream graphics cards to performance and enthusiast segments. These GPUs range from $150 at the low-end to over $1,000 for premium cards, with Nvidia’s gaming gross margins in the high 50s.

Web behemoths such as Google, Facebook, Amazon, and Microsoft have found GPUs to be adept at accelerating cloud workloads. To train a computer to recognize spoken words or images, it must be exposed to massive amounts of data to educate itself. Conclusions involve taking what the model learned during the training process and putting it into real-world applications to make decisions. The training process is ideal for GPUs that have massively parallel architecture.

Nvidia has become a key player in the artificial intelligence accelerator market with its GPUs for AI training and possibility workloads. The firm launched its latest A100 data center GPUs. The A100 boasts impressive performance enhancements from its predecessor (V100). Nvidia is the sole beneficiary of the burgeoning ai and self-driving trends. Data center revenue grew considerably, as customers leverage both Nvidia’s training and key AI applications such as natural language processing (NLP).

Google

Sun, 25 Aug 2019 00:00:00 +0000

Google uses technology to improve the user experience in nearly all its offerings while making the sale and purchase of ads efficient for publishers and advertisers. The adoption of mobile devices has been increasing. The online advertising market has taken notice and is following its target audience onto the mobile platform. Google partake in this mega-trend on the back of its Android mobile operating system’s growing market share, helping drive revenue growth and maintain Google’s leadership across many internet categories.

Google’s search engine is perceived as being the most advanced in the industry, which allows the company to remain focused on innovation and long-term growth opportunities. Google has leveraged its technical expertise to create its private cloud platform and increase its market share in this space. In the mobile internet space, Android’s dominant global market share of smartphones leaves Google well positioned to continue generating top-line growth as search traffic shifts from desktop to mobile.

Google has a massive consumer base that allows the company to collect data for precision advertising. Google can offer an attractive return on investment (ROI) for advertisers and build a growing network of advertising customers. Through network effects, the addition of each ad and advertiser improves the efficiency of Google’s programmatic advertising offerings, allowing the firm to better monetize the network.

Google has applied machine learning to its Google App, Gmail, and cloud offerings. As technological advancements improve the user experience for each product, the likelihood of further usage also increases. Google’s continuing investment in machine learning should help increase the effectiveness of ad rankings and placements. The monetization of machine learning stems from the fact that technology increases the volume and click-through rates (CTR) of ads. Google monitors real-time search trends, where ad impressions can be purchased in advance. Relatedly, programmatic video ad spending has grown at a healthy rate.

Google holds significant intangible assets related to overall technological expertise in search algorithms and machine learning, as well as access to and accumulation of data that is deemed valuable to advertisers. Google’s brand is a significant intangible asset, as Google has become eponymous with searching. The Google brand as a significant driver of user growth for YouTube, Maps, Gmail, and Chrome. An expanding user base helps the company collect more data and monetize through online ads. As Google has successfully increased its users’ dependence on its products, the company also managed to keep changing the usage of those products to increase the user switching cost.

Facebook

Fri, 16 Aug 2019 00:00:00 +0000

Facebook ’s ad revenue per user has been growing continuously, demonstrating the value that advertisers see in working with the firm. The application of AI technology to Facebook ’s various offerings, along with the launch of VR products, will increase user engagement, driving further growth in advertising revenue. Facebook provides the most massive audience and the most valuable data for social network online advertising.

The firm’s Facebook app, along with Instagram, Messenger, and WhatsApp, is among the world’s most widely used apps on both Android and iPhone smartphones. Facebook is taking steps to further monetize its various Apps, such as providing interactive video ads. The firm is also applying artificial intelligence and virtual and augmented reality technologies to various products, which may increase Facebook user engagement even further, helping to further generate attractive revenue growth from advertisers in the future.

The company’s main offerings all have strong network effects. These network effects serve to both create barriers to success for new social - network upstarts as well as barriers to exit for existing users who might leave behind friends, contacts, pictures, memories, and more by departing to alternative platforms. Facebook has emerged and established itself as the clear-cut social media leader.

Facebook has also slowly become an entertainment hub, which helps increase engagement and user time spent on Facebook. The users are posting more videos and providing a live video feed from where they are at a certain point in time. Additional apps created by developers on the Facebook platform also help maintain users within the Facebook ecosystem. Users are on Facebook and Instagram a combined 65 minutes per day (compared with a daily 64 - minute average during 2016 - 2018 ) posting videos and photos.

Unlike any other online platform globally, Facebook has accumulated data about everyone with Facebook and/or an Instagram account. Without the need for cookies enabled on desktop or mobile browsers, the firm knows its users ’ browsing history on many non-Facebook sites or apps. With access to such data and to billions of photos and videos uploaded by its users, Facebook continues to enhance the social network by offering even more relevant content. This positive feedback loop further increases the value of its data asset, which only Facebook and its advertising partners can monetize. Specifically, Facebook monetizes user information only by using it to increase the effectiveness of its advertisers. Advertisers ’ willingness to use Facebook is demonstrated by the 26 % average annual growth of Facebook ’s average ad revenue per user (ARPU), during the past five years.

Facebook ’s large and growing user base and the rich data that it generates help advertisers post more effective target ads, in terms of brand awareness, resulting in a high return on investment (ROI). Most Facebook users are now accessing Facebook and its apps via mobile devices. Thus, the main driver behind the growth of online advertising has been the mobile advertising market and video advertisers.

The firm ’s management has demonstrated its focus on long-term returns on investments. With a large amount of cash, along with little debt, Facebook is well-positioned to make additional investments in acquisitions or more research and development. The firm will continue to make decisions regarding capital allocation that are beneficial for its social network users and its shareholders.

Apple

Fri, 09 Aug 2019 00:00:00 +0000

Apple has plenty of opportunities to reap the rewards of its iPhone business. Apple’s iPhone and IOS operating systems have consistently been rated at the head of the pack in terms of customer loyalty, engagement, and security, which bodes well for long-term customer retention. Apple pay, Apple Watch, Apple TV, and Airpods are also driving incremental revenue and helping to retain iPhone users over time.

Apple’s walled garden approach for its popular iOS allows it to charge a premium for relatively commoditized hardware not too different from that sold by companies like Samsung, and Dell. Customer switching costs are elevated for Apple users as a non-apple IOS experience does not exist. Unlike computing platforms for the windows or android ecosystems that boast PCs and smartphones, Apple enjoys stellar returns on its devices by offering unique user experiences with its iOS ecosystem.

The robust App store helped foster iPhone adoption and grew Apple’s user base, with applications ranging from productivity, social media, gaming, and so on. Apple has also been focusing on newer software and services to augment the user experience and retain customers. For Apple’s customers, few other technology titans are offering comparable expertise in both hardware and software design, which allows the firm to more seamlessly build integrated products.

Competitors such as Samsung and Google specialize in hardware and software, respectively. But neither Samsung nor Google has been able to offer a comprehensive and integrated product like the iPhone. Both have attempted to develop software or operating systems (Samsung’s Tizen OS) and hardware (Google’s Pixel smartphone). Nonetheless, Apple’s expertise in both hardware and software represents an intangible asset that even the strongest of tech firms have struggled to replicate.

The active base of apple devices reached 1.5 billion at the end of 2019, up from 1. 4 billion a year prior, showing the strong stickiness Apple has created. These switching costs are not insurmountable, illustrated by the rise and fall of former mobile device titans such as Nokia, Motorola, and Blackberry. Apple may not be immune to pitfalls, as consumer sentiment for technology gadgets can be unforgiving, with one buggy or subpar product potentially driving customers to other companies.

Apple’s integration of hardware and software also supports its developer networks. Apple iOS will be loaded on to only a handful of screen sizes or iPhone models, versus the hundreds of devices and manufacturers that support android. This leads to a more fragmented Android ecosystem, which is relatively harder for developers to support. Some competitors are willing to sell hardware at essentially cost to drive revenue or stickiness in other business segments. A recent focus on AI assistants such as Google Now and Amazon Alexa has also put pressure on apple’s Siri that has fallen behind its peers in efficacy. Herein lies another area Apple may face headwinds if consumers further prioritize voice-recognition capabilities. Apple’s device may be at risk as it is not likely to supersede their iOS counterparts.

Amazon

Fri, 02 Aug 2019 00:00:00 +0000

Amazon dominates North America ’s online retail category with estimated gross merchandise volume (GMV) of approximately $275 billion in the region during 2019. With more than half of the world ’s internet users coming from developing markets, Amazon has ample international growth opportunities for its marketplaces, cloud services, advertising, and devices. Amazon’s device portfolio - including Kindle, Fire TV, Dash, Echo, and Alexa - enabled products - are intriguing customer acquisition/retention tools.

The combination of competitive pricing, unparalleled logistics capabilities and speed, and high- level customer service makes Amazon an increasingly vital distribution channel for consumer brands. Amazon maintains a consumer value proposition by expediting the Prime shipping and expanding digital content library, and new member benefits. As the traditional brick-and-mortar retail industry has undergone a period of rapid transformation over the past several years.

Amazon generates strong cash flow from seller services, which is reinvested in advertising, service, and website enhancements that keep its marketplace strong. Amazon’s brand has come to represent low prices, a wide selection, convenience, and superior customer service — a rare combination among retailers. Amazon ’s operational efficiency of the distribution network, which satisfies consumer demand for free and expedited shipping.

Amazon benefits from a network effect, as low prices, an expansive breadth of products, and a user-friendly interface attract millions of customers, which in return, attracts merchants of all kinds to amazon.com. Customer reviews, product recommendations, and wish lists increase in relevance as more consumers and products are added to the amazon platform, enhancing its network effect. The customer reviews and product recommendations are growing in presence as more retailers are adding to the site, enhancing the network effect.

Amazon has continued to develop into a formidable player in digital media, given its vast content offerings, into new verticals, and the ability to sell hardware. Amazon ’s cloud computing offerings possess more than 3 times the computing capacity in use than the next 10 largest providers combined. With investments in additional capacity and other innovations, AWS has become an increasingly positive gross margin contributor.

Alibaba

Wed, 31 Jul 2019 00:00:00 +0000

Alibaba is a big data-centric conglomerate, with transaction data from its marketplaces, financial services, and logistics businesses. Big data has also allowed the company to move into cloud computing, media/entertainment, and online-to-offline services. The strong network effect and big data obsession enable the leading E-commerce platform player in China to extend into many other growth avenues.

Alibaba’s internet services affect a staggering vast majority of Chinese internet users, an 83 % penetration rate for the Taobao/Tmall E-commerce marketplaces by the end of 2019. This provides Alibaba with an unparalleled source of data that it can use to help merchants and consumer brands develop personalized mobile marketing and content strategies to expand their target audience.

Alibaba benefits from a strong “two-sided” network effect. The value of the platform to consumers increases with a greater number of sellers, and vice versa. Millions of Chinese consumers consider Taobao and Tmall as their default go-to options when seeking products and services online. Taobao users with a strong appetite for branded products can shop Tmall for a better shopping experience, and assurance of higher quality, whereas Taobao’s online marketplaces are interconnected, which compounds its network effect then breeds other competitive advantages and growth opportunities. Taobao and Tmall have developed a powerful brand as Alibaba’s core intangible assets.

The decision to consolidate Cainiao and invest more heavily in new smart warehousing and other logistics technologies will shift Alibababa away from a pure virtual platform. The enhanced logistics capabilities stemming from its partnership with other retailers strengthen this platform’s network effect and make it more compelling for buyers and sellers.

Alibaba has had limited success with its previous e-commerce endeavors outside China. The acquisition of NetEase’s cross-border online commerce platform Kaola is considered complementary to Tmall Global. Global merchants have enjoyed a low cost and flexibility on the Alibaba’s global purchase platform, which has helped Alibaba maintain and accelerate their market share in import and outport services.

Alibaba’s desktop and mobile monetization rates have been on upward trends. The company has also increased its investments in user experience as other local and global players look to expand its presence in China. Alibaba has increased its monetization rates overtime via improved seller conversion rates from personalized the embracing of data-enriched marketing tools by mobile sellers, and increased contribution from Tmall.

Alibaba Cloud is likely to develop into a growth engine and a significant cash flow contributor overtime for the company. The early mover advantages in big data and cloud computing in China gives the company distinct advantages. Increased demand from corporations and other government groups look to reduce information technology expenditures.

An Ongoing Collection of Data Sources (Strategy Research)

Fri, 21 Jun 2019 00:00:00 +0000

This is an ongoing collection of data sources.

Industries and Economies

Companies

Internet and Patents

An Ongoing Collection of NLP Startups

Tue, 21 May 2019 00:00:00 +0000

This is an ongoing collection of NLP Startups.

Content Marketing

Bibblio (Content Recommendation)
inPowered (Content Amplification and Promotion)
IBM Watson Marketing (Digital STP; not a startup, but very cool)

Content Analysis

Frase (Content Analytics)
Market Muse (Content Analytics)
Unbounce (Content Analytics and Optimization for Landing Pages)
Mixpanel (Consumer Behavior Analytics)
Dataminr (Public Information Discovery and Detection)

Text Generation

Automated Insights (Text Generation from Data)
Persado (Text Generation with Keyword Optimization)
Phrasee (Text Generation for Copywriting)

Chatbot

Microsoft Azure Bot (Chatbot Builder; not a startup, but very cool)
Mobile Monkey (Chatbot Builder)
Live Person (Chatbot w/ Real-time Analytics)

Citation Analysis on Corporate Strategy & Network Change

Tue, 23 Apr 2019 00:00:00 +0000

1 Central Themes: Network of Co-occurrence Keywords

2 the “Classic” Papers

## [1] "most central papers (top 30) in co-citation network"

##       BAUM JAC 2000        ANJOS F 2015       GREVE H. 2014        LAVIE D 2006 
##            1.000000            1.000000            1.000000            1.000000 
##       PHELPS C 2012     PODOLNY JM 2001      PORRINI P 2004 TATARYNOWICZ A 2016 
##            1.000000            1.000000            1.000000            1.000000 
##       ZAHEER A 2010        AHUJA G 2000        ERDOS P 1959     GULATI R 1999-1 
##            1.000000            1.000000            1.000000            1.000000 
##  HASPESLAGH P. 1991    HERNANDEZ E 2015     HIGGINS MJ 2006         LIN N. 2001 
##            1.000000            1.000000            1.000000            1.000000 
##       ROGAN M. 2013        ROGAN M 2014        SYTCH M 2014       ZAHEER A 2005 
##            1.000000            1.000000            1.000000            1.000000 
##        BURT R. 1992      BUSKENS V 2008        DEVOS E 2009        AHUJA G 2012 
##            1.000000            1.000000            1.000000            1.000000 
##    BARABASI AL 1999       RYALL MD 2007      SHAVER JM 2006       ADLER N. 2005 
##            1.000000            1.000000            1.000000            0.572516 
##        ARORA A 1990         BALA V 2000 
##            0.572516            0.572516

An Ongoing Collection of Blogdown Sites

Mon, 22 Apr 2019 00:00:00 +0000

This is an ongoing collection of blogdown tutorials and websites.

Tutorials

Blogdown Documentation (by Yihui Xie, Amber Thomas, Alison Hill)
Blogdown Tutorial (by Amber Thomas)
Blogdown Tutorial (by Alison Hill)

Websites

Yihui Xie’s Website
Rob J. Hyndman’s Website
Live Free or Dichotomize (by Lucy and Nick)
Emily C. Zabor’s Website
Awesome Blogdown (a list of blogs built with blogdown)

Citation Analysis of Recent Acquisition Research

Mon, 15 Apr 2019 00:00:00 +0000

1 Sample and Data

The sample used in this analysis contains 160 articles on “mergers and acquisitions” published on top management and finance journals since 2008.

2 Central Themes

2.1 Network of Co-occurrence Keywords

This graph shows the most common themes (top 30 co-occuring keywords) of these papers.

2.2 MDS Mapping of the Conceptual Structure

We can use MDS (Multi-dimentional Scaling) and Dendrogram to map the distance/dissimilarity among the themes.

3 Central Papers

3.1 the “Main Stream” Papers

## [1] "most central papers (top 20) in bibliographic coupling network"

##   BARKEMA HG, 2008       KIM JY, 2009   NADOLSKA A, 2014 
##          1.0000000          0.8682856          0.8598660 
##       KIM JY, 2011 HALEBLIAN JJ, 2017   GORANOVA M, 2010 
##          0.8555986          0.8210859          0.7714827 
##    DEVERS CE, 2013 STEINBACH AL, 2017      KROLL M, 2008 
##          0.7253757          0.7121932          0.7104844 
##  MUEHLFELD K, 2012  EL-KHATIB R, 2015     ELLIS KM, 2011 
##          0.6669592          0.6612580          0.6551406 
##  MCDONALD ML, 2008      SHI W, 2017-1 HEIMERIKS KH, 2012 
##          0.6185902          0.5899796          0.5799630 
##  GRAEBNER ME, 2009      ZOLLO M, 2009     ELLIS KM, 2009 
##          0.5753946          0.5740604          0.5715228 
##        YIM S, 2013      ZOLLO M, 2010 
##          0.5578773          0.5558791

To identify the “main stream” papers, we can construct a “bibliographic coupling” (BC) network to see which papers co-cited the same prior research with other papers. The top 20 “main stream” (by eigenvector centrality in the BC network) papers.

3.2 the “Classic” Papers

## [1] "most central papers (top 20) in co-citation network"

##    MOELLER SB 2004   HALEBLIAN J 1999     JENSEN MC 1986 
##          1.0000000          0.9891969          0.9313744 
##       MORCK R 1990   HAYWARD MLA 2002 HASPESLAGH P. 1991 
##          0.8737119          0.8477468          0.7896179 
##   HAYWARD MLA 1997       KING DR 2004        ROLL R 1986 
##          0.7119846          0.7104482          0.6958790 
##       ZOLLO M 2004    MASULIS RW 2007  MOELLER SB 2005-1 
##          0.6931354          0.6809497          0.6665792 
##     ASQUITH P 1983     ANDRADE G 2001     JENSEN MC 1976 
##          0.6269530          0.6209498          0.5782434 
##  MALMENDIER U 2008  HALEBLIAN JJ 2006      FULLER K 2002 
##          0.5702609          0.5595749          0.5314969 
##  HECKMAN JJ 1979-1     GOMPERS P 2003 
##          0.5272799          0.5248858

To identify the “classic” papers, we can construct a “co-citation” (CC) network to see which papers are cited together with other papers by later researchers. The top 20 “classic” (by eigenvector centrality in the CC network) papers.

A.I./Blockchain Adoption and Technology Substitution

Tue, 02 Apr 2019 00:00:00 +0000

Artificial intelligence (A.I.) and blockchain are the hottest new technologies right now. It would be fun to think about the future of them (see the patterns, draw the inference, and make predictions). Would A.I. and blockchain be the next “internet,” which has disrupted many traditional business models and creates many new giants across the globe? Or, since businesses have learned a lesson from the internet disruption, A.I. and blockchain will be merely incorporated into the existing technology terrain as an expansion of opportunities by the incumbents? Do A.I. and blockchain have the same fate?

I’ve been leaving an eye on the adoption/acquisition of new technology in the banking sector (since one of my dissertation chapters was on banking acquisitions). Here is some latest news.

The banking sector has a bright prospect incorporating the A.I. in its six core functions.

JP Morgan has adopted cryptocurrency in the payment businesses.

Wells Fargo and Mastercard CEOs express doubt about blockchain.

PayPal made a major investment on the blockchain.

Could there be a unified explanation for what is happening right now? Well, one may associate the topic of technology development with Clayton Christensen’s ‘innovator’s dilemma’ assertion, which predicts emerging new technology will quickly take over declining old ones (often depicted as the double-S curves). It’s the good management–trying best to hear about customer needs, play with competition carefully, and align resource allocation with calculated risk/benefit profiles–who prevent the incumbents from adopting new technologies. Then shouldn’t managers try their best? According to Clayton Christensen, it is difficult for good management not to do good–it is difficult for the incumbents to utilize new technologies even they tend to spot them earlier than startups. At the initiation stage of new technology, incumbents can’t risk their existing market share and customer relationships with unproven models, whereas startups have little to lose. The prediction is: startups disrupt incumbents.

However, things are different right now. The boundary between new tech. and old tech. are melting down; they are intertwined ecosystems rather than isolated communities. Incumbents right now have taken deliberate effort and allocate resource for emerging technology. They are agiler and more determined to disrupt themselves before being disrupted by others. If important pre-existing conditions are different, it might be necessary to modify the predictions from Clayton Christensen’s double-S curves .

First, the compatibility between incumbents’ existing technology and emerging technology should lower the likelihood and rate of disruption. Let’s consider two cases here. One is about different incumbent technologies: as compared to traditional banking, the higher level of digitization of PayPal allows a higher compatibility potential with blockchain (consider internet and blockchain don’t rely on a central authority/agency as traditional banking does). The other is about different emerging technologies: as compared to the blockchain, A.I. allows a higher level of compatibility with traditional banking (both banking and A.I. are aimed at decision/prediction accuracy and efficiency.)

Second, the intensity of the incumbent technology expansion should also lower the likelihood of disruption. This factor is not only related to incumbents’ existing resource base but also has something to do with management. A recent paper in SMJ ¹ has updated the “double-S” framework by incorporating the “incumbent technology expansion” in predicting the rate of substitution and found empirical support for the model’s prediction efficacy.

Conclusion: since A.I. tends to have a higher level of compatibility with, and also face a stronger expansion threat from, the incumbent technology than blockchain, it’s predicted to have a lower likelihood/rate of disruption. However, contingencies should also exist due to the significant variance in incumbents’ technology, resource, and management conditions.

Adner, R., & Kapoor, R. (2016). Innovation ecosystems and the pace of substitution: Re‐examining technology S‐curves. Strategic management journal, 37(4), 625-648.↩

A Summary of Recent 'Social Evaluation' Research

Mon, 25 Mar 2019 00:00:00 +0000

1 Motivation

The goal of this post is to analyze and review the “social evaluation” research tracing back to Zuckerman (1999)¹. The analysis/review is based on hand-collected papers that directly and indirectly cite Zuckerman (1999). The review is not intended to be comprehensive but to understand the central ideas on the topic developed by management/sociology scholars in the past 20 years.

Let’s continue on the “positioning” topic from one of my previous posts (firms are social actors being evaluated by social audiences), but shift our focus of discussion from “what actors do” to “what audience think.” That is, instead of an “actor” standpoint, let’s take an “audience” perspective to consider how firms as market participants are evaluated by market audiences. Reality doesn’t directly translate into human decisions; human beings can only understand reality from their brain, which is subjected to social construction. As a result, the evaluations on market participants do not come from complete rationality in a social vacuum. Rather, audiences’ cognition is shaped by the market’s categorical structure and/or the socially accepted categorical beliefs.

2. Biblio-Analysis

2.1 Selection Criteria

## Warning: The curvature argument has been deprecated in favour of strength

The raw data is based on 29 hand-collected papers directed and indirectly citing Zukerman (1999) and extracted from Web of Science. The papers that not on the topic of “social evaluation” are excluded from the sample, based on the author’s judgment. We can get a general idea of the citation relationship among those papers from the direct citation network graph.

2.2 Central Sources and Keywords

By quickly checking the origins (papers being cited) of the sampled papers, we can see that a big portion (the blue-colored cluster) is the classic institution work. It makes sense, as the discussion of the underlying mechanism of social evaluation should be related to how the evaluating standards are formed in the institutionalization processes. The summary in 3.2 will be based on the top-cited “social evaluation” papers in the red-colored cluster.

The inter-connected keyword clusters present the relationship among the main topics covered by this stream of research. It seems to have a quite strong “strategy” flavor². The most frequently used keywords include “industry”, “market”, and “performance.”

3 Classic Papers

3.1 Top Citations

by frequency:

## [1] "ZUCKERMAN EW, 1999, AM J SOCIOL, V104, P1398, DOI 10.1086/210178"          
## [2] "HSU G, 2006, ADMIN SCI QUART, V51, P420, DOI 10.2189/ASQU.51.3.420"        
## [3] "HSU G, 2009, AM SOCIOL REV, V74, P150, DOI 10.1177/000312240907400108"     
## [4] "PONTIKES EG, 2012, ADMIN SCI QUART, V57, P81, DOI 10.1177/0001839212446689"
## [5] "RAO H, 2005, AM SOCIOL REV, V70, P968, DOI 10.1177/000312240507000605"

by co-citation centrality:

## [1] "most central sources (top 5)"

## ZUCKERMAN EW 1999        HSU G 2009      HSU G 2006-1  PONTIKES EG 2012 
##         1.0000000         0.7142248         0.6970316         0.6754593 
##        RAO H 2005 
##         0.6150981

The most frequent (by frequency count) citations are also the most central (by eigenvector centrality) citations.

3.2 Summary of Top-cited Papers

Hsu et al. (2009) provide an integrative explanation for the multi-category disadvantage. The paper combines the actor- and audience-centered explantions and argues that category spanning is both difficult for the actor to manage and hard for the audience to understand. Hsu (2006) explicitly analyzes the underlying cognition mechanism in which film audiences encounter difficulties in making sense of multi-genre films. Pontikes (2012) analyzes the different segments of market audiences with distinct preferences for ambiguous classification–the difficulty of classifying startups into established categories. According to her, consumers and corporate venture capitalists are more passive market-takers who prefer startups that are easy to classify; in contrast, individual venture capitalists are market-makers who have the power to redefine the market structure and see ambiguous classification as opportunities. Rao et al. (2005) argue that the market boundaries are melting when high-status market participants’ cross the market categories–French chefs with starts in the Guide Michelin make hybrid cuisines. Typically, there are cost and risk associated with “cross-category borrowings.” But they tend to attenuate when borrowings prevail. Overall, it seems to be risky to cross the categorical boundaries. But contextual contingencies for the benefit/risk profile of category-straddling exists.

References:

Zuckerman, E. W. (1999). The categorical imperative: Securities analysts and the illegitimacy discount. American Journal of Sociology, 104(5), 1398-1438.

Hsu, G., Hannan, M. T., & Koçak, Ö. (2009). Multiple category memberships in markets: An integrative theory and two empirical tests. American Sociological Review, 74(1), 150-169.

Hsu, G. (2006). Jacks of all trades and masters of none: Audiences’ reactions to spanning genres in feature film production. Administrative Science Quarterly, 51(3), 420-450.

Pontikes, E. G. (2012). Two sides of the same coin: How ambiguous classification affects multiple audiences’ evaluations. Administrative Science Quarterly, 57(1), 81-118.

Rao, H., Monin, P., & Durand, R. (2005). Border crossing: Bricolage and the erosion of categorical boundaries in French gastronomy. American Sociological Review, 70(6), 968-991.

The central idea of this paper is that firms’ stock price is affected by the categorical pattern of financial analysts’ industry coverage, which reflects the institionalized expectations for firms’ product mix straddling across various market categories.↩
For the central topics in strategy research, refer to “the intellectual structure of the strategic management field…” by Nerur et al. (2008) and “what is strategic management…” by Nag el al. (2007)↩

Interoperability in Tech. Evolution—Thoughts on “Reticulate"

Mon, 11 Mar 2019 00:00:00 +0000

“In our highly dynamic world, it’s not enough for an organization to possess a competitive advantage at a point in time; it needs an evolutionary advantage over time—a capacity to change as fast as change itself; to change before a crisis breaks.” — Gary Hamel

On March 6th, the R package “reticulate”(1.11.1) was released on Cran, which allows R users to directly call Python objects or use Python in an R interface.¹ Bloggers have been praising this development for the ground-breaking interoperability between R and Python. With “reticulate,” we no longer need to jump between different IDEs to leverage the complementarity between R and Python. This will certainly bring big ease to the programming routine of many “bilingual” data scientist using both R and Python.

As an occupational habit, I tend to think about the value creation and value capture in the dynamics of industry and technology evolution. Here are some thoughts on the “reticulate” case.

First of all, one the user level, “reticulate” by Rstudio will benefit those who program in both R and Python. So far, R and Python are the two most popular programming languages in data science. A decision for a newcomer in data science to make is always: “R vs. Python, which one should I choose?” Now this headache is significantly lessened. If we can overcome the “language barrier” easily, we won’t be stuck in the either-or any more.

Second, on the community level, there have already been quite a few early adopters who are R-Python bilinguals even before the pre-release of “reticulate.” They choose a language depending on which one is more readily usable for a specific problem. For bilinguals, there is no transition in their head, but there are some inefficient transitions on their fingertips. Now, they will leverage the power of “interoperability” to keep their hand movements in one place–Rstudio. These users will be engaged in Rstudio more often. What will also naturally happen? (a) the total number of R-Python bilinguals will grow; (b) the two languages will become more integrated, instead of being divided. We will see fewer (or slower growth of) packages/modules which have the same name and perform the same functionality in two languages, and also less “R vs Python comparisons.”

Third, thinking from the technological (and also strategic) perspective, we see the greatly increased compatibility between two programming languages. What follows compatibility will be an increase in the adoption rate on the technology who opens up its door to embrace other technologies. Rstudio is the superpower in the R domain. Such a move will draw many Python programmers to R and also attract and retain R users with the empowerment of R-Python interoperability. This “one-way” compatibility² will greatly enhance the usability and desirability of the Rstudio as a platform provider.

The release of “reticulate” is a strategic move of Rstudio. More data scientists will be stay engaged in Rstudio, or they will spend more fo their time on the platform. But this is not the end of the story. The interoperability will continuously enlarge the overlap between the R and Python communities if we draw a Venn diagram in our mind. The Python-centered IEDs are also powerful. They have great features that their users wouldn’t live without. They may match the move of Rstudio. More likely, the reverse “recuticate” in Python will emerge from the large open source community, by innumerable open source developers. Even if the one-way compatibility will continue, Python IEDs will not lose the game. As R programmers can easily operate Python in Rstudio, they are also more likely to travel and contribute to the Python world. This is similar to the case when Apple allows Microsoft’s Windows operating system to be installed on Mac and Amazon’s Kindle electronic book app installed on iPad. The competing platforms will both benefit from the one-way compatibility, because the enlarged overlap of the Venn diagram will bring “two-way” flows of traffic to both sides.³

Open source can always provide more flexibility to the users, developers, and businesses involved. Perhaps we cannot formulate the dynamics of technological evolution in a game-theoretic fashion, as the players in open source are not “competing” in the same way. For example, Anaconda, the company who distributes many Python IDEs (including JupyterLab and Spyder), has long been providing Rstudio IDE on their cloud and GUI. I see “reticulate” as a strategic, but also very harmonious move to create net benefits to the open source community for data science at large.

This can be easily done in Rstudio. Webinar by Rstudio Documentation by Rstudio ↩︎
The existing python IDEs have not provided an interface for R-Python interoperability yet. Python programmers tend to use JupyterLab, Rodeo, Spyder, Visual Studio Code, and PyCharm, none of which have such an interoperability feature right now.↩︎
For a more detailed analysis: Adner, R., Chen, J., & Zhu, F. (2016). Frenemies in Platform Markets: The Case of Apple’s iPad vs. Amazon’s Kindle. Harvard Business School Technology & Operations Mgt. Unit Working Paper, (15-087).↩︎

“MMeM”: Modeling the Multivariate Mixed-effects

Sun, 10 Mar 2019 00:00:00 +0000

I co-developed an R pakcage “MMeM” for estimating the variance-covariance matrix of random effects $\mathbf{u}$ and $\mathbf{e}$ on multiple dependent variables.

Statistical Intuition

The univariate mixed-effects model

In univariate mixed-effects model: $\mathbf{y} = \mathbf{Xb} + \mathbf{Zu} + \mathbf{e}$ (1), we estimate the variance component $\sigma_u^2$ and $\sigma_e^2$ for $\mathbf{u} \sim N(\mathbf{0}, \sigma_u^2\mathbf{I})$ and $\mathbf{e} \sim N(\mathbf{0}, \sigma_e^2\mathbf{I})$.

In formula (1): $\mathbf{y}$ is n $\times$ 1 response vector; $\mathbf{X}$ and $\mathbf{Z}$ are n $\times$ p and n $\times$ s; $\mathbf{b}$ is p $\times$ 1 coefficients vector for the fixed effects; $\mathbf{u}$ is s $\times$ 1 matrix for the random effects, $\mathbf{e}$ is n $\times$ 1 vector of random errors.

The multivariate mixed-effects model

In multivariate mixed-effects model: $\mathbf{y} = (\mathbf{I} \otimes \mathbf{X})\mathbf{b} + (\mathbf{I} \otimes \mathbf{Z} )\mathbf{u} + \mathbf{e}$ (2), in which $\mathbf{y} = \left\{\mathbf{y}_i\right\}_c, \mathbf{b} = \left\{\mathbf{b}_i\right\}_c, \mathbf{u} = \left\{\mathbf{u}_i\right\}_c, \mathbf{e} = \left\{\mathbf{e}_i\right\}_c, i =1, \dots, q$, we estimate the variance-covariance matrix of random effects $\mathbf{u}$ and $\mathbf{e}$ on $q$ response variates, namely $\mathbf{T}$ and $\mathbf{E}$, for $var(\mathbf{u}) = \mathbf{G} = \mathbf{T}\otimes \mathbf{I}_s, var(\mathbf{e}) = \mathbf{R} = \mathbf{E} \otimes \mathbf{I}_n, var(\mathbf{y}) = \mathbf{V} = \mathbf{T}\otimes \mathbf{ZZ}' + \mathbf{E} \otimes \mathbf{I}_n$.

In formula (2): $\mathbf{y}$ is nq $\times$ 1 response vector; $\mathbf{X}$ is n $\times$ p design matrix for the fixed effects; $\mathbf{b}$ is pq $\times$ 1 coefficients vector for the fixed effects; $\mathbf{Z}$ is n $\times$ s design matrix for the random effects; $\mathbf{u}$ is sq $\times$ 1 vector of the random effects; $\mathbf{e}$ is nq $\times$ 1 vector of random errors.

An Implementation Example

# install.packages("MMeM")
library(MMeM)
data(simdata)

Initialization

# initialize with a positive-definiate var-cov
T.start = matrix(c(10, 5, 5, 15), 2, 2)
E.start = matrix(c(10, 1, 1, 3), 2, 2)

Estimation

# using the Henderson3 estimation mothod
results_henderson = MMeM_henderson3(fml = c(V1,V2) ~ X_vec + (1|Z_vec), data = simdata, factor_X = TRUE)

## Bivariate response: V1 and V2

print(results_henderson)

## $T.estimates
##                 T: V1   T: V1 V2     T: V2
## T.estimates  65.47395   9.969188  9.766204
## T.variance  807.56303 128.824836 20.552160
## 
## $E.estimates
##                E: V1  E: V1 V2     E: V2
## E.estimates 55.74506 11.477502 41.171927
## E.variance  11.09826  8.196899  6.054027

“regrrr”: One-stop R-toolkit for Compiling Regression Results

Wed, 06 Mar 2019 00:00:00 +0000

In strategy/management research, we always need to compile the regression results into the publishable format and sometimes plot the moderating effects. Thus, I developed this “regrrr” package to help do the job.

Here is the quickstart guide.

Installation

To install from CRAN:

install.packages("regrrr")
library(regrrr)

You can also use devtools to install the latest development version:

devtools::install_github("raykyang/regrrr")
library(regrrr)

Examples

compile the correlation table

library(regrrr)

## Registered S3 methods overwritten by 'tibble':
##   method     from  
##   format.tbl pillar
##   print.tbl  pillar

data(mtcars)
m0 <- lm(mpg ~ vs + carb + hp + wt, data = mtcars)
m1 <- update(m0, . ~ . + wt * hp)
m2 <- update(m1, . ~ . + wt * vs)
cor.table(data = m2$model)

##          Mean  S.D.     1     2    3    4    5
## 1.mpg   20.09  6.03  1.00                     
## 2.vs     0.44  0.50  0.66  1.00               
## 3.carb   2.81  1.62 -0.55 -0.57 1.00          
## 4.hp   146.69 68.56 -0.78 -0.72 0.75 1.00     
## 5.wt     3.22  0.98 -0.87 -0.55 0.43 0.66 1.00

compile the regression table

regression_table <- rbind(
combine_long_tab(to_long_tab(summary(m0)$coef),
                 to_long_tab(summary(m1)$coef),
                 to_long_tab(summary(m2)$coef)),
compare_models(m0, m1, m2))
rownames(regression_table) <- NULL
print(regression_table)

##        Variables   Model 0   Model 1   Model 2
## 1    (Intercept) 35.435*** 48.157*** 46.698***
## 2                  (2.503)   (4.097)   (9.272)
## 3             vs     1.353     1.077     2.171
## 4                  (1.382)   (1.152)   (6.320)
## 5           carb    -0.057    -0.043    -0.009
## 6                  (0.449)   (0.374)   (0.426)
## 7             hp   -0.024† -0.113***   -0.107*
## 8                  (0.014)   (0.027)   (0.044)
## 9             wt -3.792*** -8.071***   -7.594*
## 10                 (0.658)   (1.307)   (3.016)
## 11         hp:wt             0.027**    0.025†
## 12                           (0.008)   (0.015)
## 13         vs:wt                        -0.367
## 14                                     (2.081)
## 15     R_squared     0.833     0.889     0.889
## 16 Adj_R_squared     0.808     0.867     0.862
## 17       Delta_F            12.517**     0.031

plot the moderating effect

plot_effect(reg.coef = summary(m2)$coefficients, data = mtcars, model = m2,
            x_var.name = "wt", y_var.name = "mpg", moderator.name = "hp",
            confidence_interval = TRUE,  CI_Ribbon = FALSE, 
            xlab = "Weight", ylab = "MPG", moderator.lab = "Horsepower") +
ggplot2::theme(text=ggplot2::element_text(family="Times New Roman", size = 16))

plot the moderating effect with a linear spline

library(lspline)
data(mtcars)
m3 <- lm(mpg ~ vs + carb + hp + lspline(wt, knots = 4, marginal = FALSE) * hp, data = mtcars)
plot_effect(reg.coef=summary(m3)$coefficients, data = mtcars, model = m3, 
            x_var.name = "wt", y_var.name = "mpg", moderator.name = "hp",
            xlab="Weight", ylab="MPG", moderator.lab="Horsepower") +
ggplot2::theme(text=ggplot2::element_text(family="Times New Roman", size = 16))

As we can see from the last line of code, the plot is customizable using “ggplot2”. There are a couple of other functions. Please see the reference manual on R documentation for details.

I’m also aiming to expand the package’s usage around its core functions. If you have any ideas or want to report a bug, please contact me or suggest on the GitHub page.

System Mapping in Business Modeling

Thu, 28 Feb 2019 00:00:00 +0000

In the past several years, I spent quite some time coaching college students in the business major on their individual/team projects. These include business model design (BUS146), strategic audit (BUS109), and marketing plan (BUS103). In each quarter, there are around 100 students enrolled in BUS103/BUS109, and 50 in BUS146. And they typically select different companies and different entrepreneurship ideas to work on.

At some point, I realized it is crucial for them to quickly “model” a business so they can have a very clear picture of how businesses work. I also usually suggest they put all their thoughts on one paper. So they can have the big picture without missing out any important information. This is where “system mapping” can come in handy.

It involves three simple steps. First, brainstorm the essential components in a business system. Second, identify the causal linkages and loops. Third, mark the positive/negative relationships (and the leverage in the system). I find that most students can master this skill very quickly and can generate a good graph with some practice.

I also push for simplicity, as I notice the more I ask them to succinctly define the components, the more they move away from perceptual experiences to critical thinking. Such a map can be a helpful starting point for us to keep questioning ourselves on further facts, proofs, and underlying mechanisms.

Here is an example of a generic “precision online marketing” model (e.g., Google, Facebook, Youtube, etc.), made with iGraph.

A Quick Bibliometric Analysis on “Positioning”

Tue, 26 Feb 2019 00:00:00 +0000

I had a whim and wanted to do some quick bibliometric analysis on the topic of “strategic positioning”.

The raw data is extracted from Web of Science. The search parameters:

TITLE: ("strategic positioning" OR "industry positioning" OR "market positioning")
Refined by: WEB OF SCIENCE CATEGORIES: ( MANAGEMENT OR BUSINESS OR ECONOMICS ) AND WEB OF SCIENCE CATEGORIES: ( MANAGEMENT OR BUSINESS OR INTERNATIONAL RELATIONS OR ECONOMICS ) AND DOCUMENT TYPES: ( ARTICLE ) AND LANGUAGES: ( ENGLISH )
Timespan: All years. 
Indexes: SCI-EXPANDED, SSCI, A&HCI, CPCI-S, CPCI-SSH, BKCI-S, BKCI-SSH, ESCI, CCR-EXPANDED, IC.

I used the R package “bibliometrix” to conduct (1) “bibliographic coupling” and (2) “co-citation” analysis.

Bibliographic coupling analysis allows us to explore the “confluence and interactions” among recent papers. Two articles are bibliographically coupled if they co-cited at least one article.

Co-citation analysis allows us to trace back to the origin of thoughts and classic papers. Two articles have a co-citation tie if they are both cited in a third article.

Reference:

Aria, M. & Cuccurullo, C. (2017) bibliometrix: An R-tool for comprehensive science mapping analysis, Journal of Informetrics, 11(4), pp 959-975, Elsevier.

Optimal Positioning for Firms and Individuals

Mon, 25 Feb 2019 00:00:00 +0000

Imagine you are invited to a party. You want to get noticed by other partygoers. Or you want to be recognized as a “cool” person by those who you are interested in. The “coolness” will result in some social benefits you desire. This is a scene of “social evaluation,” in which you want to stand out of the crowd of peer actors (competitors) to impress the audiences (customers and suppliers). If you succeed, the peer actors will admire you, the audiences will adore you, and of course, the esteems and profitable relationships will follow. Now the question is, “how to position yourself to be ‘cool’ so that you can win the game?”

Practically, the first question coming to your mind might be “to what extent should I be similar to or different from others?” You want to be different so that you can stand out. However, you don’t want to be too different so that people think you are wired. Your goal is to “impress” but not to “surprise.” It is a bad idea to violate the audience’s expectations on your appearances and behaviors – the commonly accepted “norms.” Violations typically lead to penalties.

Excessive differentiation is risky. It jeopardizes your “legitimacy,” the potential for being accepted by the audience, who controls the resources and opportunities. Now the headache you get is — being similar places you under competitive pressures (so you can’t stand out) but being different gives you the risk of losing legitimacy. Then, you may find a simple solution: find a middle ground between the two extremes — “to be different as ‘legitimately’ possible.” Good. You got a “holistic” approach, in which you consider all the audiences, attributes, and evaluation processes as a “whole.” And you have a clear goal—maximizing the total audience perceptions. It is implementable. It is the wisdom of Confucius’ as well as Aristotle’s “Golden Mean.”

If you don’t want to stop here, let’s think further. We can do some analysis¹, by moving away from the “holistic” approach to an “analytical” one. For example, we can anatomize the positioning problem in three ways: (1) separating the evaluating audiences, (2) separating the actor attributes, and (3) separating the evaluation processes.

First, different audience groups (government, customers, investors, etc.) may demand different levels of conformity vs. differentiation, so you can deploy specific positioning strategy to target specific audiences (e.g. horizontal websites tend to give different people different things while vertical websites tend to meet specific people’s specific needs). Second, as audiences allocate their decision weights among multiple dimensions of difference, you can conform on most of them to maintain legitimacy and then differentiate on a couple of others to look cool (e.g. in a party, people always care about dress code while worrying about outfit clash). Third, as the audiences typically engage in a categorization process before making the final selection, you can strategize for “entering the game” and “winning the game” separately (e.g. audiences tend to make a short list first, select the winner from it. No matter you want to design a website or find a job, to make it to the shortlist, you need to first figure out the similarities that help you stay relevant; and in the selection stage, find the differences that help you win).

The positioning problem shouldn’t stop here. As audiences evaluate actor attributes with their subjective minds, numerous factors (cognitive, sociological, cultural ones) determine how individual opinions aggregate and emerge as market behavior. Researchers have been studying the patterns, drawing inferences and making predictions. Importantly, there could exist positive and negative feedback loops for the mutual influence between actor behavior (strategic action) and audience perception (performance evaluation). This is what I’m interested in.

References:

Deephouse, D. L. 1999. To be different, or to be the same? It’s a question (and theory) of strategic balance. Strategic Management Journal, 20(2): 147–166.

Zuckerman, E. W. 2016. Optimal distinctiveness revisited: An integrative framework for understanding the balance between differentiation and conformity in individual and organizational identities. Handbook of Organizational Identity.

Zhao, E. Y., Fisher, G., Lounsbury, M., & Miller, D. 2017. Optimal distinctiveness: Broadening the interface between institutional theory and strategic management. Strategic Management Journal, 38(1): 93–113.

Merriam-Webster has a definition of “analysis” — “separation of a whole into its component parts.”↩︎

A List of Market Trackers and Economy Indicators

Fri, 01 Feb 2019 00:00:00 +0000

FRED

BEA

FED

BLS

Census

Treasury

Graph

Other

Links to Quizzes

Mon, 01 Jan 0001 00:00:00 +0000

Section 024 (2:10~3:00 pm): Go to Quiz

Section 026 (4:10~5:00 pm): Go to Quiz

My Research

Mon, 01 Jan 0001 00:00:00 +0000

Check My C.V.

Yang's Strategy Blog

Mon, 01 Jan 0001 00:00:00 +0000

Hi! I'm Dr. Yang, a business professor at Tongji University, teaching digital strategy, business-model innovation, and organization theories.

My research contexts involve emerging technologies, corporate strategy, and venture capital. Beyond the phenomenon itself, I delve into the interplay between environment-shaping strategies and performance-shaping environments. C.V.

As we envision and observe how things connect, our minds turn to what makes them tick, producing knowledge in a loop of causal inference and probabilistic prediction.