econometrics - Causal inference for tabular data
Analyzes tabular data for causal effects, including diagnostics, cleaning guidance, estimator selection, robustness checks, and applied research design.
Tags
Updated: 2026-10-01Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Diagnose tabular datasets
- Recommend identification strategies
- Construct propensity scores
- Estimate treatment effects
- Run OLS, IV, DID, and RDD
- Assess robustness and falsification
- Analyze treatment-effect heterogeneity
- Interpret estimates and assumptions
Inputs
- Tabular dataset
- Outcome variable
- Treatment variable
- Covariates
- Identification strategy
- Panel structure
- Research question
Outputs
- Causal effect estimates
- Dataset diagnostic reports
- Model comparison tables
- Overlap, balance, and pre-trend diagnostics
- Robustness and falsification results
- Identification and research-design memos
- Plain-language interpretations
Requirements
- Python 3.10+
- numpy, pandas, matplotlib, statsmodels, linearmodels, scipy
- openpyxl for .xlsx/.xlsm files
- xlrd for legacy .xls files
- Access to the skill library