development-economics

What problem does it solve?

This Skill helps you turn development-economics research data into credible causal estimates, even when attrition and covariate imbalance threaten validity.

Core Features & Use Cases

• Intention-to-treat (ITT) analysis with randomization strata to estimate average treatment effects from RCTs.
• Covariate balance diagnostics using standardized mean differences and love plots to validate randomization quality.
• Selective attrition sensitivity with Lee (2009) sharp bounds, plus propensity score matching (PSM) for observational-style comparisons and FGT poverty indices (P0/P1/P2) for poverty measurement.
• Use case: You collected an LSMS-style household survey with an RCT offer (Z), partial compliance (D), and attrition (missing outcomes); you want ITT and treatment-effect bounds, assess baseline balance, and compute poverty measures such as headcount and poverty gap.

Quick Start

Ask an AI to load your RCT dataset and run an ITT regression with strata fixed effects, compute covariate balance (SMD + love plot), apply Lee sharp bounds for selective attrition, then estimate PSM ATT and calculate FGT poverty indices from baseline income.