Clustered standard errors

<p>Clustered standard errors (or Liang-Zeger standard errors) are measurements that estimate the <a href="/facts/Standard_error/Iv96AyCD">standard error</a> of a <a href="/facts/Regression_analysis/n6z5Tf7K">regression</a> parameter in settings where observations may be subdivided into smaller-sized groups ("clusters") and where the sampling and/or treatment assignment is correlated within each group. Clustered standard errors are widely used in a variety of applied econometric settings, including <a href="/facts/Difference_in_differences/Mfk1MtIJ">difference-in-differences</a> or experiments.
</p><p>Analogous to how <a href="/facts/Heteroscedasticity-consistent_standard_errors/IrTRLgVp">Huber-White</a> standard errors are <a href="/facts/Consistent_estimator/HAKDWYEM">consistent</a> in the presence of <a href="/facts/Heteroscedasticity/msBQWI7G">heteroscedasticity</a> and <a href="/facts/Newey%25E2%2580%2593West_estimator/rg22f5tu">Newey–West</a> standard errors are consistent in the presence of accurately-modeled <a href="/facts/Autocorrelation/dm4MqK0A">autocorrelation</a>, clustered standard errors are consistent in the presence of cluster-based sampling or treatment assignment. Clustered standard errors are often justified by possible correlation in <a href="/facts/Errors_and_residuals/Ef0LgAS4">modeling residuals</a> within each cluster; while recent work suggests that this is not the precise justification behind clustering, it may be <a href="/facts/Pedagogy/t1LIfLeC">pedagogically</a> useful.
</p>

Clustered standard errors open-in-new

Clustered standard errors