In s3alfisc/fwildclusterboot: Fast Wild Cluster Bootstrap Inference for Linear Models

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

Academic Papers

• Roodman, MacKinnon, Nielsen & Webb - Fast and Wild Bootstrap Inference (Stata Journal), 2019. The paper to read if you think that the performance of fwildclusterboot resembles black magic. Introduces boottest - the Stata software package fwildclusterboot is modeled after - and contains a great introduction to (almost) all features of the wild cluster bootstrap implemented in fwildclusterboot.

• MacKinnon, Nielsen & Webb - Fast and Reliable Jackknife and Bootstrap Methods for Cluster-Robust Inference (Journal of Applied Econometrics), 2023 Introduces the "31", "33", "13" bootstrap types, explains how to compute them quickly, and contrasts their empirical performance in Monte Carlo studies. Argues in favour of the "31" method.

• MacKinnon - Fast cluster bootstrap methods for linear regression models (Econometrics and Statistics, 2021) Discusses computational tricks for speeding up wild cluster bootstrap inference. Further provides a nice discussion of (bootstrap) test inversion to compute confidence intervals.

• MacKinnon, Nielsen & Webb - Cluster-robust inference: A guide to empirical practice (Journal of Econometrics), 2023 Broad introduction and state-of-the-art literature survey of concepts around clustered errors.

• Webb - Reworking wild bootstrap based inference for clustered errors (forthcoming at Canadian Journal of Economics) Introduces "Webb" weights, which are the recommended wild bootstrap weight type when the number of clusters are very small.

• MacKinnon & Webb - The wild bootstrap for few (treated) clusters (Econometrics Journal), 2018 Introduces the subcluster bootstrap for regressions with few treated clusters (e.g. difference-in-differences regressions with one treated cluster).

• MacKinnon - Wild Cluster Bootstrap Confidence Intervals (L'Actualité économique), 2015 Discusses how to invert a bootstrap to obtain a confidence interval.

• Davidson & MacKinnon - Wild bootstrap tests for IV regression (Journal of Economic and Business Statistics), 2010 Introduces the WRE bootstrap / wild cluster bootstrap for instrumental variables regression.

• Cameron, Gelbach & Miller - Bootstrap-based improvements for inference with clustered errors (Review of Economics and Statistics) The paper that started the literature on wild cluster bootstrap inference. Simulation evidence that the wild cluster bootstrap works remarkably well when there are only few clusters.

• Flachaire - Bootstrapping heteroskedastic regression models: wild bootstrap vs. pairs bootstrap (Computational Statistics & Data Analysis), 2005 Provides simulation evidence of the performance of heteroskedastic wild bootstrap procedures.

• MacKinnon - Thirty Years of Heteroskedasticity-Robust Inference, 2012 More simulation evidence on the performance of the HC1-HC3 vcov estimators vs the wild bootstrap.

• Joshi, Megha, James E. Pustejovsky, and S. Natasha Beretvas - "Cluster wild bootstrapping to handle dependent effect sizes in meta‐analysis with a small number of studies." (Research Synthesis Methods), 2022 Nice simulations on the empirical performance of the wild cluster bootstrap for tests of multiple joint hypotheses.

• Kline & Santos - A Score Based Approach to Wild Bootstrap Inference (Journal of Econometric Methods), 2012 Introduces a score based wild bootstrap for non-linear regression models.

Links, blogposts, etc

Stata blog - Heteroskedasticity robust standard errors: Some practical considerations Extensive simulations on small sample properties of HC estimators, inlcuding the wild bootstrap

s3alfisc/fwildclusterboot documentation built on Sept. 17, 2023, 5:55 a.m.