Heteroskedasticity-Robust Standard Errors for Fixed Effects Panel Data Regression

By James H. Stock, Mark W. Watson. 2008

708 citations recorded by the source corpus.

The conventional heteroskedasticity-robust (HR) variance matrix estimator for cross-sectional regression (with or without a degrees of freedom adjustment), applied to the fixed effects estimator for panel data with serially uncorrelated errors, is inconsistent if the number of time periods T is fixed (and greater than two) as the number of entities n increases. We provide a bias-adjusted HR estimator that is (nT)1/2 -consistent under any sequences (n, T) in which n and/or T increase to %u221E.The conventional heteroskedasticity-robust (HR) variance matrix estimator for cross-sectional regression (with or without a degrees of freedom adjustment), applied to the fixed effects estimator for panel data with serially uncorrelated errors, is inconsistent if the number of time periods T is fixed (and greater than two) as the number of entities n increases. We provide a bias-adjusted HR estimator that is (nT)1/2 -consistent under any sequences (n, T) in which n and/or T increase to %u221E.

Subfields: Economics and Econometrics

View Paper: https://www.semanticscholar.org/paper/55f4b9d196800b6931635bfee2237b3a83cbe5f6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

Citations in the corpus

Common Errors: How to (and Not to) Control for Unobserved Heterogeneity (2014, background): “See Wooldridge (2010, Chapters 10 and 20), Arellano (1987), and Stock and Watson (2008), for more details.”
Egalitarianism and International Investment (2011, positive): “Also, as Stock and Watson (2008) have shown, with the use of clustering we are able to control for serial autocorrelation most effectively.”
Globalization and economic growth: Evidence from two decades of transition in CEE (2014, methodology): “19 To be precise, we used the robust Huber/White/sandwich VCE estimator discussed in Wooldridge (2009), Stock and Watson (2008) and Arellano (2003). present the results of estimating growth models only for those cases in which Xt did not contain any two elements of the set {EDU, GC, M2}.”
Spillovers in Space: Does Geography Matter? (2016, positive): “An earlier version of this work contained standard errors and confidence bands based on the Pinkse, Slade, and Brett (2002) method, again using a compromise between the extremes of full clustering and Stock and Watson (2008).”
Punctuated Generosity (2013, methodology): “To account for multiple observations per firm, we used cluster-adjusted, heteroskedasticity-robust standard errors (Stock and Watson, 2008).”
When Should We Use Unit Fixed Effects Regression Models for Causal Inference with Longitudinal Data? (2019, methodology): “Second, taking advantage of this equivalence, we can use various modelbased robust standard errors for within-unit matching estimators (e.g., White, 1980a; Stock and Watson, 2008).”
Cluster–Robust Variance Estimation for Dyadic Data (2013, neutral): “With ﬁxed N and growing T ( d ) , however, additional assumptions are needed for consistency (Hansen, 2007; Stock and Watson, 2008).”