All notable changes to InsightStack are documented here.
econometrics/module fully built with 7 implementations:difference_in_differences.py— DiD estimation with parallel trends test, covariate controls, and visualizationpropensity_score_matching.py— PSM with nearest-neighbor matching, balance assessment, and ATT estimationregression_discontinuity.py— Sharp RDD with bandwidth selection, McCrary density test, and visualizationinstrumental_variables.py— 2SLS estimation with first-stage F-test diagnosticssensitivity_analysis.py— Rosenbaum bounds, placebo tests, permutation inferencedid_analysis.R— DiD in R with event study plot using ggplot2psm_analysis.R— PSM in R using MatchIt package with balance diagnostics
econometrics/sample_program_data.csv— 20-row simulated program evaluation dataset
- README updated: econometrics module status changed from Planned to Ready
impact_evaluation/: DiD, propensity score matching, ITT/LATE estimation, event study, MDE power calculationregression_discontinuity.doandheterogeneous_effects.dofor causal inference- GitHub Action for automatic changelog generation
ROADMAP.mdanddesign_coverage.md
graphs_viz/: added boxplot_by_group, panel_line_income, combine_graphs- README enhancements with badges and ecosystem links
- Core data folders:
data_management/,descriptive_analysis/,regression_models/ - Templates:
annotated_do_file.do,project_folder_template.zip - Synthetic datasets:
education_labels.csv,livelihoods_panel.csv,health_survey.csv
- Label inconsistencies in
logistic_models.do - Folder alignment and README formatting
- Project bootstrapped with InsightStack architecture
- GitHub Pages, README, and folder layout