Generates rigorous panel and causal-inference code in Python using `pyfixest` (modern fixest port), `linearmodels.PanelOLS`, `statsmodels`, and pandas, with proper fixed effects, clustering, weak-IV diagnostics, event studies, and DIME-aligned reproducibility. Honest about Python's smaller modern-DiD ecosystem and routes the user to R when needed. Use when the user asks for PanelOLS, pyfixest, fixed effects, two-way clustering, IV2SLS, panel DiD, event studies, weak instruments, BJS imputation, or publication-ready Python regression output.
Generates rigorous, modern, reproducible R code for causal inference and panel econometrics with `fixest`, heterogeneity-robust DiD estimators (Callaway-Sant'Anna, Sun-Abraham, BJS, de Chaisemartin-D'Haultfoeuille), weak-IV-robust inference, optimal-bandwidth RDD via `rdrobust`, and wild cluster bootstrap. Use when the user asks for IV, DiD, event studies, RDD, TWFE, staggered treatment, clustered or wild-bootstrap inference, instrumental variables, parallel trends, first-stage F, AR confidence sets, or publication-ready R regression output.
Generates rigorous, reproducible Stata regression workflows that follow World Bank DIME Analytics conventions (`ietoolkit`, `iefolder`, master do-files, `ieboilstart`, dynamic absolute paths, `iebaltab`, `ieddtab`, `esttab`). Defaults to modern estimators (`reghdfe`, `ivreg2`/`ivreghdfe`, `csdid`, `eventstudyinteract`, `did_imputation`, `boottest`, `rdrobust`) with weak-IV-robust inference and wild cluster bootstrap when needed. Use when the user asks for OLS, logit, fixed effects, panel regression, DiD, IV, RDD, event study, balance tables, esttab/outreg2, reghdfe, csdid, ivreg2, boottest, ietoolkit, iefolder, master do-file, or publication-ready Stata output.
Builds academic economics presentations in LaTeX Beamer that share the same `notation.tex` macros and `tabs/`/`figs/` outputs as the underlying paper, so a single rebuild keeps slides and paper in sync. Defaults to clean themes (`metropolis` / `default` with custom colors), 16:9 aspect ratio, time-budget-aware structure (15/20/45/60/90 min), incremental reveals, and DIME-aligned reproducibility (figures/tables produced by code, never screenshotted). Cross-links to `latex-econ-model`, `latex-tables`, `econ-visualization`, and `academic-paper-writer`. Use when the user asks for conference talk slides, seminar presentations, job-market talks, thesis defense slides, or wants to convert a paper into Beamer.
Generates publication-quality economics figures produced by code (R `ggplot2`, Python `matplotlib`/`seaborn`, Stata `twoway`/`coefplot`) and exported in vector format directly to the paper's `figs/` folder. Defaults to DIME's "full replicability" tier and the [Reviewing Graphs checklist](https://dimewiki.worldbank.org/Checklist:_Reviewing_Graphs) — clear titles for standalone use, intuitive colors, colorblind-safe palettes, consistent axis labels, source citations on standalone visuals, and visualization choices grounded in the [Data Visualization](https://dimewiki.worldbank.org/Data_visualization) wiki page. Use when the user asks for event-study coefficient plots, balance plots, time series with recession shading, choropleth maps, binscatters, density plots, scatter-with-fit, regression coefficient plots, or any reproducible figure for a paper, slide deck, or dashboard.
Generates rigorous, reproducible Python data-acquisition pipelines for economic APIs (FRED, World Bank, BLS, OECD SDMX, IMF, Eurostat, Yahoo Finance). Applies DIME Analytics conceptual conventions to Python: DataWork-style folders with immutable Raw/, master orchestration script with dynamic absolute paths via `pathlib`, secrets-as-env-vars (PII discipline applied to API keys), source codebooks recording series IDs / units / vintage / access date, deterministic caching, and asserted output schemas. Use when the user asks to download economic indicators, build cross-country panels, automate data refreshes, combine multiple API sources, set up a reproducible data pipeline, or document data provenance.
Generates rigorous, reproducible Stata data-cleaning workflows that follow World Bank DIME Analytics conventions (`iefieldkit`, `ietoolkit`, `iefolder` DataWork structure, `ieboilstart`, `iecodebook`, `ieduplicates`, `iecompdup`, extended missing values, master datasets, encryption for PII, dynamic absolute paths). Defaults to codebook-driven cleaning, `assert`/`isid` validation, labeled categorical variables, and reproducible from-clean-session execution. Use when the user asks to clean survey or administrative data in Stata, build an analysis-ready panel, handle duplicates, harmonize datasets across rounds, write a master dataset, de-identify data, label variables, or produce a codebook.
Foundational data-discipline skill for econometrics work in any language. Enforces the discipline that every other data, analysis, and writing skill depends on: define the unit of analysis before anything else; treat IDs as a contract; merge with explicit cardinality and validation; never silently change the unit of observation; protect PII; document every drop; keep raw immutable; build a master dataset per unit of observation. Cross-language patterns for Stata, R, and Python; aligned with DIME Analytics' [Data Cleaning](https://dimewiki.worldbank.org/Data_Cleaning), [ID Variable Properties](https://dimewiki.worldbank.org/ID_Variable_Properties), and [DataWork Folder](https://dimewiki.worldbank.org/DataWork_Folder) guidance. Use when the user asks to combine datasets, build a panel, validate data structure, set up a project's data folder, handle PII / de-identification, define or change the unit of analysis, debug a merge, or whenever multiple data skills are about to interact (e.g. before running an analysi