A curated list of our technical postings (World Bank) for econometrics issues; Recent Applied Micro Econometrics techniques
Stata Coding Guide, Clean multiple Excel Files; Stata and Python Translation; Python and Econometrics
Making tables from Stata and from R; visualisation with R; Stata for Undergrad
Even Study with R; Deep Learning for Economists; ML for Data Science (with Ethics); Causal Inference for Stats
Undergraduate Econometrics with R (from Science Po's Department of Economics). Statisticall Inference and Regression; Time Series (Forecasting) with R; Mixtape Sessions (collection of applied econometrics short courses)
Undergraduate level's computational economics with Python, and advanced level courses; Python and Data Science for Public Policy from LSE and Chicago; Data Scaping (Duke), Text Analysis,
Data Science for Economists (University of Oregon); web-scraping
Master-level Prediction and ML (Good course for Year 3 UK, this book, a version found here, is good, more techniques)
Testing the significance of a product of two coefficients (Or, equivalently, Ho: beta1 = 0 or beta2 = 0). This is an interesting question in the meditation analysis literature. See this for a review of conventional tests and why they won't work. See this for a new approach. The nlcom command in Stata.
Interaction terms for logit/probit models. Be careful.; testing one-sided multiple hypothesis? Use an LR test.
PhD's training course: https://github.com/paulgp/applied-methods-phd
A collection of various statistical methods with different statistical software: https://lost-stats.github.io/Machine_Learning/Machine_Learning.html
Stata to Latex guide: https://medium.com/the-stata-guide/the-stata-to-latex-guide-6e7ed5622856
Exogenous IV? https://www.stata-journal.com/article.html?article=st0538
Exercises for Machine Learning courses: https://arxiv.org/pdf/2206.13446.pdf, and a classic textbook for probablistic Machine Learning; OLS and ML by Cameron
Most exciting developments in statistics in the last 50 years?
Sampling-based versus design-based uncertainty in regression analysis (Abadie et al., 2020, Econometrica): discusses how to calculate the standard error (sampling uncertainty) under the design-based uncertainty (randomness comes from the assignment of the treatment, instead of random sampling). This is useful for applications where a population of interest is observed in its entirety. Another approach to using the "super-population" argument for when we observe the entire population but report standard errors.
Inflation and Interest Rate Targets - a nice summary from the NBER
Another fundamental read from Abadie et al. 2022 (ReStud) - why do we use clustered standard error?
Taking logs when the dependent variable has many zero values. (NBER working paper)
Difference in Differences Literature; What to do if parallel trend fails?
Theory papers: Designing interrogation (ReStud, 2024); and Misinformation Online (ReStud, 2024); the Economics of Bitcoin (QJE, 25)
Lecture Notes: Game Theory (Caltech); and an X thread on many maths courses
Robustness with OVB: changing signs and attenuating: https://arxiv.org/abs/2208.00552
The Economics of (Gen)AI Collection: (i) Machine Translation improves American-Spanish trades on Ebay (MS, 2019); (ii) GenAI in research (LSE, 2025); (iii) GenAI and Work Productivity (QJE, 2025), more references here (iv) brief intro in GenAI (Google, 2025), GenAI Hallucination and Skills; JEP