Causal Inference Workbench DiD · event study · synthetic control · regression discontinuity

Three panel-data estimators that economists reach for when a policy hits some units and not others: two-way fixed-effects difference-in-differences with cluster-robust standard errors, an event study (plain TWFE or the Sun–Abraham interaction-weighted version for staggered adoption), and Abadie-style synthetic control with placebo-in-space inference. Everything runs in this tab. Nothing is uploaded anywhere.

1 · Data

A panel with unit and year effects, an AR(1) shock, and a treatment effect you choose. Because the truth is known, you can see which estimator recovers it.

Long format: one row per unit-period. A 0/1 treatment column that switches on when a unit is treated, or a column with each unit's adoption period (blank or 0 for never treated).

Pulls a country-year panel straight from the World Bank API (api.worldbank.org). Aggregates are dropped. You name the treated country and the year.

2 · Estimation settings

Panel

No data yet. Simulate a panel, upload one, or fetch one.

Regression discontinuity local linear · triangular kernel · IK bandwidth

A separate estimator with its own data: an outcome, a running variable and a cutoff. Eligibility thresholds (a BPL score, a land ceiling, an age or marks cutoff) make this the natural design in India.

What is being estimated

τ = limx↓c E[y|x] − limx↑c E[y|x], estimated by two local linear regressions, one on each side of the cutoff, with triangular kernel weights that fall to zero at the bandwidth h (Hahn, Todd and van der Klaauw 2001). Standard errors are heteroskedasticity-robust (HC1) from the two weighted fits; this is the “conventional” interval, not the bias-corrected one of Calonico, Cattaneo and Titiunik (2014), which the page does not implement. The default bandwidth follows Imbens and Kalyanaraman (2012): a pilot bandwidth gives the density at the cutoff and the conditional variances on each side, cubic and then quadratic fits give the second derivatives, a regularisation term guards against tiny curvature, and the plug-in constant for the triangular kernel is 3.4375. The bandwidth-sensitivity plot re-estimates τ from half to twice that bandwidth. The McCrary (2008) test bins the running variable, fits the density on each side of the cutoff with a local linear regression on the bin heights, and tests whether the log density jumps; a jump means units are sorting across the cutoff and the design is compromised. Placebo cutoffs re-run the estimator at the median of each side, where no effect should appear.

Need help setting out your story, what sets your work apart, the context, the return on investment and the projections? Pinpoint Ventures can help. Write to hello@pinpointventures.in or message on WhatsApp.