Kweku Opoku-Agyemang, Ph.D
@KwekuOA
Followers
9,280
Following
6,037
Media
2,139
Statuses
52,697
CEO @mlxdoing AI. @DevEconX The next generation. Aff. @The_IGC . Ex-postdoc @UCBerkeley , @cornell_tech . PhD @UWMadison @ohiou 👉
#EconAI 👆 + #DevEcon 👉
Joined January 2009
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
North Carolina
• 927844 Tweets
Hurricane Helene
• 703802 Tweets
#FreeMorara
• 363975 Tweets
FEMA
• 353514 Tweets
Dikembe Mutombo
• 144249 Tweets
Mets
• 92424 Tweets
Mets
• 92424 Tweets
Game 2
• 87140 Tweets
Diaz
• 87109 Tweets
Mimi
• 74593 Tweets
Verizon
• 74450 Tweets
#النصر_الريان
• 53264 Tweets
#الاهلي_الوصل
• 44024 Tweets
#KYSvBJK
• 33349 Tweets
Kemp
• 31067 Tweets
#KızılGoncalar
• 23944 Tweets
Kayseri
• 22956 Tweets
Onana
• 22571 Tweets
Semih
• 18761 Tweets
الجيش اللبناني
• 16546 Tweets
رونالدو
• 14880 Tweets
Lindor
• 13352 Tweets
Gavin Creel
• 11750 Tweets
Bournemouth
• 11293 Tweets
Southampton
• 10945 Tweets
Ozzie
• 10555 Tweets
Last Seen Profiles
Pinned Tweet
“What are you working on” has never been a more exciting question for me than right now.
Welcome to the new era of industrial research and academic-level products that blur the lines between academia and industry for mutual benefit.
A thread.
2
30
64
Data science for economists: version control, web scraping, spatial analysis, workflow, project management and other needs.
8
254
1K
A gentle introduction of economists to the Python programming language. These notes should help you to be able to use it efficiently and independently. An approach to data through the lens of economics discipline:
4
154
907
Job market tip:
“Don’t act like a grad student.”
What I mean is, don’t be asking for approval. Nobody wants to think about why you fit there. Decide you fit and tell them why. Act more like you’re already in the role you want. Sound like you already work there.
3
60
856
A lesson in game theory in development economics...
It’s either the moped or the bag. Your choice.
1K
37K
294K
8
123
828
Draw all your economics diagrams in LaTeX without tears:
0
148
734
A gentle introduction to Python for economists. Individuals with an approach to data through the economic discipline wishing to discover Python would do well to check this out.
4
113
709
According to the most conservative estimates in this QJE, the introduction of the potato accounts for approximately one-quarter of the growth in Old World population and urbanization between 1700 and 1900.
1
160
668
Another thread on classic industrial R&D labs. Bell Labs was the birthplace of information theory. Nine Nobel prizes(!) and four Turing awards(!!). I’m mostly quoting this neat new .
@Nature
article on how they operated with actual quotes.
8
140
607
"Amazon has hired more than 150 PhD economists, making it the largest employer in the field behind the Federal Reserve...according to...Amazon itself, integrating economists has been critical to the company’s phenomenal growth in e-commerce."
8
103
544
A programming guide for economists, regardless of language, to:
• Simplify data & program management
• Maintain accuracy & efficiency
• Ensure reproducibility
• Keep a workflow organized throughout the project's lifecycle
1
107
499
Data science for economists is a game-changer! 🚀 Master version control, web scraping, spatial analysis, workflow, project management, and more.
Don’t miss out—level up your skills now:
3
130
615
With the Mathpix app, you can convert actual images of math to LaTeX:
Just take a screenshot of the math equations and paste into your editor, all with a single keyboard shortcut.
It also works with hand-written equations(!)
The Mathpix app allows users to take screenshots of
#math
equations and paste the extracted Latex
11
163
412
7
141
463
Recent advances in applied micro and econometrics. A reading list:
2
95
394
Neat statistics lecture notes to introduce you to causal inference: from randomized controlled trials (RCTs) to adaptive experiments.
1
90
391
Python materials for economists, this time with the datasets to accompany the slides:
Slides:
Datasets:
0
79
383
@mischiefanimals
The mountain lion is like:
*Record scratch*
*freeze frame*
"Yup, that's me. You're probably wondering how I ended up in this situation."
0
12
352
An important review in Nature argues that scientists need more time to think. "That suggests an even more radical rethink of the current funding model for academic research."
7
106
330
Causal inference with distribution functions means questions like:
How does X affect the *distribution* of income or purchases?
Brand-new work in JRSS:B has the Rubin Causal Model and a new Econometrica has synthetic controls.
I made the staggered DID 👉
2
76
323
New methods at the interface of ML and econometrics perform better than either traditional ML or econometric methods when applied to causal inference for average treatment effects, optimal policy estimation, & estimation of the counterfactual effect:
1
59
314
Learn how to think mathematically with a guided journey into the world of abstraction and writing proofs.
📒Everything you always wanted to know about mathematics (*but didn’t even know to ask).
A cool 😎 free👏 book (700 pages).
0
74
277
How to avoid machine learning pitfalls: a guide for academic researchers
1
46
269
A JEP introduction to event study models. One of the most appealing features of event study models is that it provides a built-in graphical summary of results, which can reveal rich patterns of behavior.
0
73
266
A First Course in Causal Inference: these >400 page lecture notes only require basic knowledge of probability theory, statistical inference, and linear and logistic regressions. From the Yule-Simpson paradox through potential outcomes, mediation and more:
0
59
263
This repository tracks the developments in Difference-in-Difference (DiD) software packages:
4
65
262
An introduction to event study models—a powerful tool used for the purpose of estimating dynamic treatment effects in natural and randomized experiments. It provides a built-in graphical summary of results, which can reveal rich patterns of behavior.
1
57
252
Even math majors can get low quant GRE scores, which may seem mysterious. I'll partly unpack the phenomenon for the young people in the house. A thread on how to get a perfect or near-perfect math score (in many cases) on the GRE without tears.
7
24
247
Causal inference: What if:
Full text:
Code in SAS, Stata, R, Python, Julia and Parametric g-formula software in R and SAS:
1
80
250
Notes on randomized experiments, regression discontinuity designs, matching, regression, instrumental variables, difference-in-differences, and dynamic causal models; from political science, economics, education, public health, and other disciplines.
0
73
245
Machine learning for economics research: when, what and how: a gentle introduction out of the Bank of Canada:
1
56
234
Dividing 1 by 998,001 will give you all three-digit numbers from 000 to 999 in order, except for 998! 😄
0
47
239
A First Course in Causal Inference: these >400 page lecture notes only require basic knowledge of probability theory, statistical inference, and linear and logistic regressions. From the Yule-Simpson paradox through potential outcomes, mediation and more:
2
62
227
Unlock the power of the microeconomic toolkit! 📈 Researchers, lecturers, and practitioners (real estate, etc.)—make your thinking crystal clear.
Don’t miss out on drawing flawless economics diagrams in LaTeX without tears:
2
54
260
An AER finds: relative to the widely used difference-in-differences and synthetic control methods—both theoretically and empirically—a “synthetic difference-in-differences” estimator has desirable robustness properties, performs well in common settings:
0
37
207
This Econometrica is how one of the largest multinationals of the 20th century—the United Fruit Company—had a positive and persistent effect on living standards, because, to attract & maintain the workforce, the firm invested heavily in education & health:
7
41
205
Dynamic Structural Econometrics. This is a treasure-trove of resources, from the 2019-2024 Econometric Society Summer Schools.
YouTube videos:
Slides, Notes, Readings, Code:
Go!
1
56
210
A First Course in Causal Inference: these >400 page lecture notes only require basic knowledge of probability theory, statistical inference, and linear and logistic regressions. From the Yule-Simpson paradox through potential outcomes, mediation and more:
1
45
203
This AER will upgrade your data visualizations focusing on binscatter including novel visualization tools, principled covariate adjustment, estimation of conditional mean functions, and more for your Python, R and Stata needs.
1
43
194
What is the rate of return on everything? Is the aggregate real rate of return in the economy higher than the growth rate of the economy and, if so, by how much? With evidence from 1870 to 2015, this QJE reveals many new findings and puzzles:
2
54
193
Does monetary policy affect inequality? While different households respond to changes in monetary policy for different reasons, this JEP argues that the overall consumption effects are relatively evenly distributed across households.
1
47
188
The “percentogram”—a histogram binned by percentages of the cumulative distribution, rather than using fixed bin widths
1
36
191
A QJE model in which spatial differences in job loss emerge because of systematic differences between employers across local labor markets accounts for over 75% of the cross-sectional dispersion in French unemployment rates:
2
46
180
This QJE finds that by providing Black servicemembers a stable and well-paying army job and by opening doors to higher-paid postservice employment, the army significantly closes the Black-white earnings gap in the sample.
5
40
166
Also useful for young economists:
Here's the handout I use in my math courses to coach my students on good mathematical writing:
#iteachmath
28
435
2K
1
31
165
A QJE paper finds the labor market increasingly rewards social skills. Between 1980 and 2012, jobs requiring high social interaction grew by ~12 percentage points as a share of US labor. Math-intensive less social jobs—fell by 3.3 percentage points:
1
39
156
In economics, the equivalent is probably Nick Cox of Durham University, who seems to have answers to every question under the sun concerning STATA and has bailed out countless economists and social scientists over the years.
3
21
155
For economists who need a fast but beautiful map—The open source Uber AI platform is a autonomous visualization tool too few on
#EconTwitter
know about—just upload your geocoded files, add special effects you want and go!
3
36
154
My Mum at Harvard Law talking about education and its opportunities:
1
13
150
Congrats to David Card for his economics Nobel Prize: a great day for both Canada and UC Berkeley economics!
2
18
146
A JEP introduction to event study models. One of the most appealing features of event study models is that it provides a built-in graphical summary of results, which can reveal rich patterns of behavior.
1
44
145
This post on
#DataScience
for economists has resonated with many. Let me expand on why this intersection is crucial and how we're applying these principles in two exciting ventures. 🧵👇
#EconTwitter
#Innovation
#AI
"
Data science for economists: version control, web scraping, spatial analysis, workflow, project management and other needs.
8
254
1K
3
21
142
A lesson in game theory for the economics of education and administration:
1
26
136
A JEP introduction to event study models. One of the most appealing features of event study models is that it provides a built-in graphical summary of results, which can reveal rich patterns of behavior. Dive in:
1
27
136
This QJE finds: those who benefited from social spending from the New Deal were markedly more patriotic during World War II: they bought more war bonds, volunteered more, and, as soldiers, won more medals.
1
35
133
Tech economists are getting to the top of the industry. Congrats to Elizabeth Stone (an economics PhD in finance and macroeconomics who was at Lyft) for becoming the new CTO of Netflix!
2
23
131
Tips on how to run surveys: a guide to creating your own identifying variation
(by .
@S_Stantcheva
)
1
36
133
200+ South African firefighters landed in Edmonton, Canada to assist in the fight against the raging wildfire. 🇿🇦🇨🇦
2
19
128
A machine learning-based causal inference tutorial in R: causal panel data and matrix completion methods, from two-way fixed effects, interactive fixed effects, synthetic control methods, and synthetic difference-in-differences and more:
1
31
132
An AER finds: relative to the widely used difference-in-differences and synthetic control methods—both theoretically and empirically—a “synthetic difference-in-differences” estimator has desirable robustness properties, performs well in common settings:
0
27
130
Tip. Overleaf has a LaTeX Review Response Template for responding to reviewer's comments:
0
23
129
Check out this cool paper (JPE R&R), on a simple method outperforming differences-in-differences with shorter confidence intervals, and being robust to parallel trend violations, with an R package.
Paper:
Code:
3
26
130
Identifying corruption via typography: Evidence from Calibri
4
43
123
A meta-essay for economists on writing better referee reports: Learn tips and tricks for writing referee reports in economics.
1
22
129
Big data is changing economics. The largest firms in the US are valued chiefly for their data. Yet, these data are excluded from macroeconomic & finance research. New tools for measuring economic activity, market power, data markets & financial markets:
1
27
121
Great paper by .
@oyebolaoo
conditionally accepted at AEJ: Applied: Participants randomly assigned to do national service in a Nigerian state with a different ethnic majority from them are 5X more likely to live all across the country 7 years(!) later.
4
30
122
How important is agglomeration (productivity gains from clustering production and workers) for aggregate growth? Local agglomeration has an economically and statistically significant impact on the growth rate of per capita consumption, by about 10 percent.
1
30
118
This AER will upgrade your data visualizations focusing on binscatter including novel visualization tools, principled covariate adjustment, estimation of conditional mean functions, and more for your Python, R and Stata needs.
0
32
117
Research suggests that online echo chambers are a myth.
So why might social media seem to polarize users?
This PNAS paper suggests that it is not through isolation, but conflict. Interesting premise:
0
43
114
“Behavioral economics is challenging because it takes economists out of their comfort zone, but economists have to be uncomfortable.”
1
48
116
The estimator in this Econometrica relies on the same theory as changes-in-changes & works for both repeated cross-sections & panels with as little as a single pre-treatment period. Also a unique counterfactual quantile function for any distribution type.
1
15
114
A book written in 1910 could teach you calculus:
Full text:
1
24
115
Louder for people in the back. (Courtesy of Instagram.)
1
22
113
Macroeconomics is another area with machine learning integration: an early work is Emi Nakamura's (2005) paper using neural networks for inflation forecasting: A more recent overview of the literature is in Coulombe et al (2019)
4
30
112
Researchers partition large networks into clusters. For asymptotically independence to hold, it is necessary and sufficient that clusters have low conductance, (the ratio of edge boundary size to volume). This yields a simple measure of cluster quality:
1
19
113
Unlike textbook firms, focusing on physical tangible capital (machines), labor (workers), and know-how: Today many firms, have almost no physical capital. Their capital are “intangible”: software, data, design, reputation, supply-chain expertise, and R&D.
1
16
108
Thinking about applying synthetic controls to estimate causal effects in comparative case studies with aggregate units? If you have access to data within units, you can use Distributional Synthetic Controls to obtain an entire counterfactual distribution:
1
20
108
A Harvard economics student Wesley Wang spent 2+ years making a short film in high school and posted it online. Darren Aronofsky saw it & emailed him. After a bidding war, TriStar Pictures bought the rights. Wang will adapt it to film to be produced by Aronofsky’s Protozoa firm.
1
7
107
This paper provides an overview of the methods used for algorithmic text analysis in economics; defining four core empirical tasks that encompass most text-as-data research in economics, and enumerates the various approaches taken so far.
0
26
105
The age distribution for civilian personnel and scientists working on the Manhattan Project:
0
16
101
An AER finds: relative to the widely used difference-in-differences and synthetic control methods—both theoretically and empirically—a “synthetic difference-in-differences” estimator has desirable robustness properties, performs well in common settings:
1
25
106
@Andy_VGC
It's a testament to just how strong the culture at Nintendo is. The attrition rate is super-low, and the creativity has been consistent for decades. A future challenge at Nintendo will be creating new genre-defining franchises, but their cultural system is clearly proven.
3
6
101
This great AEJ: EP paper presents evidence of the positive causal impacts of research and development (R&D) tax incentives on a firm’s own innovation and that of its technological neighbors (spillovers).
1
31
103
Macroeconomic development remains because of its ability to lift entire populations out of poverty. This JEL emphasizes a synthesis of methods and ideas, with the science of experiments as a unifying feature. The synthesis is underway:
1
23
99
In 1961, some believed most jobs could be gone by 1971.
5
33
97
The Best of Both Worlds: Combining Randomized Controlled Trials with Structural Modeling: (thread):
1
22
96
This Econometrica RCT provided firms with different amounts of information on the economy, to isolate the causal effect of uncertainty. Higher macroeconomic uncertainty causes firms to reduce their prices, employment, and investment relative to prior plans
1
39
98
A Stata package to estimate panel event studies, xtevent 2.2.0, is now available in Github and in SSC!
Now with estimation in staggered adoption contexts with Sun and Abraham (2021)'s method:
To install, visit: or type -ssc install xtevent- from Stata.
0
21
96
Japan's mastery of train travel is facilitated by subtle behavioral tricks:
1
16
97
Tip. Overleaf has a new LaTeX Review Response Template for responding to reviewer's comments:
0
23
97
JEL overview: Thanks to increases in both the amount of data available and computing power, it has become possible to exploit recent formal methods to infer the statistical properties of unobservable variables from multiple imperfect measurements. Thread:
1
23
96