pith. sign in

arxiv: 1901.01869 · v2 · pith:5DROYW3Hnew · submitted 2019-01-07 · 📊 stat.AP

Patterns of Effects and Sensitivity Analysis for Differences-in-Differences

classification 📊 stat.AP
keywords biasconfoundersmethodremovesanalysiscomparisondevelopdifferences-in-differences
0
0 comments X
read the original abstract

Applied analysts often use the differences-in-differences (DID) method to estimate the causal effect of policy interventions with observational data. The method is widely used, as the required before and after comparison of a treated and control group is commonly encountered in practice. DID removes bias from unobserved time-invariant confounders. While DID removes bias from time-invariant confounders, bias from time-varying confounders may be present. Hence, like any observational comparison, DID studies remain susceptible to bias from hidden confounders. Here, we develop a method of sensitivity analysis that allows investigators to quantify the amount of bias necessary to change a study's conclusions. Our method operates within a matched design that removes bias from observed baseline covariates. We develop methods for both binary and continuous outcomes. We then apply our methods to two different empirical examples from the social sciences. In the first application, we study the effect of changes to disability payments in Germany. In the second, we re-examine whether election day registration increased turnout in Wisconsin.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Stochastic Sensitivity Analysis for Matched Observational Studies

    stat.ME 2026-06 unverdicted novelty 7.0

    Stochastic sensitivity analysis for matched studies finds worst-case conditional laws for hidden confounders instead of worst-case realizations, controlled by a sensitivity parameter that permits imperfect alignment w...

  2. Optimal Treatment Policy Estimation for Recurrent Events with a Competing Terminal Event: An Instrumented Difference-in-Differences Approach

    stat.ME 2026-06 unverdicted novelty 7.0

    Develops a multiply robust iDID estimator for optimal policies in recurrent events with terminal competing risk, with simulation results and application to Medicare Type 2 diabetes data.