In Econometrics, '''Huber-White standard errors''' are standard errors that are adjusted for correlations of error terms across observations, especially in panel and survey data as well as data with cluster structure. This type of adjusted errors is also called Sandwich, robust or empirical standard errors. Once obtained, these estimated errors should be used instead of traditional standard error estimates for inferences and hypothesis testing of the econometric model. ==References== *{{cite journal | author=Huber, P. J. | title="The behavior of maximum likelihood estimates under non-standard conditions" | journal=Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability | year=1967 | volume=4 | pages= 221-233. }} *{{cite journal | author=White, H. | title="A heteroskedasticity-consistent covariace matrix estimatorand: a direct test for heteroscedasticity" | journal=Econometrica | year=1980 | volume=48 | pages= 817-830 }} *Nawar, Abdel-Hameed (2005) "Seemingly Unrelated Regressions (SUR): An Introduction by Examples," A Statistics Lecture Note, New York University, July, mimeo *{{cite book | author=Greene, William. | title="Econometric Analysis" | publisher=Prentice Hall | year=1998 }} {{econometrics-stub}} Category:Econometrics Category:Simultaneous_equation_methods_(econometrics) Category:Mathematical_and_quantitative_methods_(economics)