Placebo inference on treatment effects when the number of clusters is small
From MaRDI portal
Publication:2330752
Abstract: I introduce a general, Fisher-style randomization testing framework to conduct nearly exact inference about the lack of effect of a binary treatment in the presence of very few, large clusters when the treatment effect is identified across clusters. The proposed randomization test formalizes and extends the intuitive notion of generating null distributions by assigning placebo treatments to untreated clusters. I show that under simple and easily verifiable conditions, the placebo test leads to asymptotically valid inference in a very large class of empirically relevant models. Examples discussed explicitly are (i) least squares regression with cluster-level treatment, (ii) difference-in-differences estimation, and (iii) binary choice models with cluster-level treatment. A simulation study and an empirical example are provided. The proposed inference procedure is easy to implement and performs well with as few as three treated and three untreated clusters.
Recommendations
- Randomization inference for difference-in-differences with few treated clusters
- On the implementation of approximate randomization tests in linear models with a small number of clusters
- Inference with difference-in-differences revisited
- Covariate balance in simple, stratified and clustered comparative studies
- Randomization Inference in a Group–Randomized Trial of Treatments for Depression
Cites work
- t-Statistic Based Correlation and Heterogeneity Robust Inference
- A NEW ASYMPTOTIC THEORY FOR HETEROSKEDASTICITY-AUTOCORRELATION ROBUST TESTS
- A Smoothed Maximum Score Estimator for the Binary Response Model
- A central limit theorem for stationary random fields
- Asymptotic Statistics
- Asymptotic properties of a robust variance matrix estimator for panel data when \(T\) is large
- Asymptotic spectral theory for nonlinear time series
- Conditional rank tests for the two-sample problem under random censorship
- HETEROSKEDASTICITY-AUTOCORRELATION ROBUST TESTING USING BANDWIDTH EQUAL TO SAMPLE SIZE
- How Much Should We Trust Differences-In-Differences Estimates?
- Inference with dependent data using cluster covariance estimators
- Nonlinear system theory: Another look at dependence
- OLS Estimation in a Model Where a Microvariable is Explained by Aggregates and Contemporaneous Disturbances are Equicorrelated
- On the Theory of Some Non-Parametric Hypotheses
- Probability. Theory and examples.
- Randomization tests under an approximate symmetry assumption
- Resampling Student's \(t\)-type statistics
- Some results on the convergence of conditional distributions
- Studentized permutation tests for non-i.i.d. hypotheses and the generalized Behrens-Fisher problem
- The Large-Sample Power of Tests Based on Permutations of Observations
- Weak dependence. With examples and applications.
Cited in
(7)- Cluster-robust inference: a guide to empirical practice
- Wild bootstrap inference for instrumental variables regressions with weak and few clusters
- Placebo Response as a Latent Characteristic: Application to Analysis of Sequential Parallel Comparison Design Studies
- Some impossibility results for inference with cluster dependence with large clusters
- Simple and trustworthy cluster-robust GMM inference
- Randomization inference for difference-in-differences with few treated clusters
- On the implementation of approximate randomization tests in linear models with a small number of clusters
This page was built for publication: Placebo inference on treatment effects when the number of clusters is small
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2330752)
