Efficient robust estimation of time-series regression models. (English) Zbl 1189.62140

Summary: This paper studies a new class of robust regression estimators based on the two-step least weighted squares (2S-LWS) estimator which employs data-adaptive weights determined from the empirical distribution or quantile functions of the regression residuals obtained from an initial robust fit. Just like many existing two-step robust methods, the proposed 2S-LWS estimator preserves robust properties of the initial robust estimate. However, contrary to the existing methods, the first-order asymptotic behavior of 2S-LWS is fully independent of the initial estimate under mild conditions. We propose data-adaptive weighting schemes that perform well both in cross-section and time-series data and prove the asymptotic normality and efficiency of the resulting procedures. A simulation study documents the theoretical properties in finite samples.


62M10 Time series, auto-correlation, regression, etc. in statistics (GARCH)
62F12 Asymptotic properties of parametric estimators
62F35 Robustness and adaptive procedures (parametric inference)
62L12 Sequential estimation
65C60 Computational problems in statistics (MSC2010)
Full Text: DOI EuDML


[1] N. S. Balke, T. B. Fomby: Large shocks, small shocks, and economic fluctuations: outliers in macroeconomic time series. J. Appl. Econom. 9 (1994), 181–200.
[2] P. Čížek: Least trimmed squares in nonlinear regression under dependence. J. Statist. Plann. Inference 136 (2006), 3967–3988. · Zbl 1103.62061
[3] P. Čížek: General trimmed estimation: robust approach to nonlinear and limited dependent variable models. CentER discussion paper. Econom. Theory. To appear.
[4] P. Čížek: Efficient robust estimation of regression models. CentER discussion paper Vol. 87. CentER. Tilburg University, Tilburg, 2007.
[5] M. G. Genton, A. Lucas: Comprehensive definitions of breakdown points for independent and dependent observations. J. R. Stat. Soc., Stat. Methodol. Ser. B 65 (2003), 81–94. · Zbl 1063.62038
[6] D. Gervini, V. J. Yohai: A class of robust and fully efficient regression estimators. Ann. Stat. 30 (2002), 583–616. · Zbl 1012.62073
[7] L. Mašíček: Diagnostics and sensitivity of robust models. Unpublished Ph.D. Thesis. Faculty of Mathematics and Physics, Charles University, Prague, 2004.
[8] A. Preminger, R. Franck: Foreign exchange rates: a robust regression approach. Int. J. Forecasting 23 (2007), 71–84.
[9] P. J. Rousseeuw: Multivariate estimation with high breakdown point. Mathematical statistics and applications, Vol. B (W. Grossman, G. Pflug, I. Vincze, W. Wertz, eds.). Reidel, Dordrecht, 1985, pp. 283–297. · Zbl 0609.62054
[10] P. J. Rousseeuw, A. M. Leroy: Robust Regression and Outlier Detection. John Wiley & Sons, New York, 1987. · Zbl 0711.62030
[11] S. Sakata, H. White: High breakdown point conditional dispersion estimation with application to S&P 500 daily returns volatility. Econometrica 66 (1998), 529–567. · Zbl 1015.91530
[12] J. R. W. Temple: Robustness tests of the augmented Solow model. J. Appl. Econometrics 13 (1998), 361–375.
[13] D. Van Dijk, P. H. Franses, A. Lucas: Testing for ARCH in the presence of additive outliers. J. Appl. Econom. 14 (1999), 539–562.
[14] J. Á. Víšek: The least weighted squares II. Consistency and asymptotic normality. Bulletin of the Czech Econom. Soc. 9 (2002), 1–28.
[15] J.Á. Víšek: Instrumental weighted variables. Austr. J. Stat. 35 (2006), 379–387.
[16] J. Woo: Economic, political, and institutional determinants of public deficits. J. Public Economics 87 (2003), 387–426.
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.