Abstract
Regression model with fixed and random effects estimated by modified versions of the Ordinary Least Squares (OLS) is a standard tool of panel data analysis. However, it is vulnerable to the bad effects of influential observations (contamination and/or atypical observations). The paper offers robustified versions of the classical methods for this framework. The robustification is carried out by the same idea which was employed when robustifying OLS, it is the idea of weighting down the large order statistics of squared residuals. In contrast to the approach based on the M-estimators this approach does not need the studentization of residuals to reach the scale- and regression-equivariance of estimator in question. Moreover, such approach is not vulnerable with respect the inliers. The numerical study reveals the reliability of the respective algorithm. The results of this study were collected in a file which is possible to find on web, address is given below. Patterns of these results were included also into the paper. The possibility to reach nearly the full efficiency of estimation - due to the iteratively tailored weight function - in the case when there are no influential points is also demonstrated.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
Amos CI (1994) Robust variance-components approach for assessing genetic linkage, in pedigrees. Am J Hum Genet 54:535–543
Atkinson AC, Riani M (2000) Robust diagnostic regresion analysis. Springer, New York
Bickel PJ (1975) One-step Huber estimates in the linear model. JASA 70:428–433
Bramati MC, Croux C (2007) Robust estimators for the fixed effects panel data model. Econ J 10:521–540
Boček P, Lachout P (1993) Linear programming approach to LMS-estimation. Memorial vol Comput Statist & Data Anal 19(1995):129–134
Balestra P, Nerlove M (1966) Source pooling cross section and time series data in the estimation of a dynamic model: The demand for natural gas. Econometrica 34:585–612
Cameron AC, Gelbach JB, Miller DL (2011) Robust inference with multiway clustering. J Bus & Econ Stat 29:238–249
Cragg JG (1983) More efficient estimation in the presence of heteroscedasticity of unknown form. Econometrica 51:751–763
Croux C, Rousseeuw PJ (1992) A class of high-breakdown scale estimators based on subranges. Comm Statist- Theory Methods 21:1935–1951
Davies L (1990) The asymptotics of S-estimators in the linear regression model. Ann Statistics 18:1651–1675
Dehon C, Gassner M, Veradi V (2009) Beware of ‘good’ outliers and overoptimistic conclussios. Oxford Bull Econ Statist 71:437–452
Fisher RA (1920) A mathematical examination of the methods of determining the accuracy of an observation by the mean error and by the mean squares error. Mon Not Roy Astr Soc 80:758–770
Galton F (1886) Regression towards mediocrity in hereditary stature. J Anthropol Inst 15:246–263
Gill PS (2000) A robust mixed linear model analysis for longitudinal data. Stat Med 19:975?987
Hampel FR, Ronchetti EM, Rousseeuw PJ, Stahel WA (1986) Robust statistics – the approach based on influence functions. Wiley, New York
Hausman J (1978) Specification test in econometrics. Econometrica 46(1978):1251–1271
Hausman JA, Taylor WE (1981) Panel data and unobservable individual effects. Econometrica 49:1377–1398
Hettmansperger TP, Sheather SJ (1992) A cautionary note on the method of least median squares. Am Stat 46:79–83
Huber PJ (1981) Robust statistics. Wiley, New York
Jones DR (1992) Meta-analysis of observational epidemiological studies: A review. J R Soc Med 85:165–168
Judge GG, Griffiths WE, Hill RC, Lutkepohl H, Lee TC (1985) The Theory and Practice of Econometrics. Wiley, New York
Jurečková J, Sen PK (1984) On adaptive scale-equivariant M-estimators in linear models Statistics and Decisions 2 (1984), Suppl. Issue No. 1.
Koenker R, Bassett G (1978) Regression quantiles. Econometrica 46:33–50
Koenker R (2004) Quantile regression for longitudinal data. J Multivar Anal 91:74–89
Kott P (1989) Robust small domain estimation using random effects modelling. Survey Methodology 15:1–12
Maddala GS (1971) The use of variance components models in pooling cross section and time series data. Econometrica 39:341–358
Pinheiro JC, Liu C (2001) Efficient Algorithms for Robust Estimation, in Linear mixed-Effects models using the multivariate t-Distribution. J Comput Graph Stat 10:249–276
Portnoy S (1983) Tightness of the sequence of empiric c.d.f. processes defined from regression fractiles. In: Franke J, Ha̋rdle W, Martin D (eds) Robust and nonlinear time-series analysis. Springer-Verlag, NY 1983, pp 231–246
Rocke DM (1991) Robustness and balance in the mixed models. Biometrics 47:303–309
Rousseeuw PJ (1984) Least median of square regression. JASA 79:871–880
Schall R (1991) Estimation in generalized linear models with random effects. Biometrika 72:593–9
Šindelář J (1994) On L-estimators viewed as M-estimators. Kybernetika 30 (1994):5,551–562
Telser LG (1964) Iterative Simultaneous Estimation of Sets of Linear Regressions. J Am Stat Assoc 59:845–62
Veradi V, Croux C (2009) Robust regression in Stata. Stat J 9(3):439–453
Veradi V, Wagner J (2010) Robust estimation of of linear fixed effects panel data model with an application to the exporter productivity premium Discussion paper No 4928, IZA, Bonn
Víšek JÁ (1996) A cautionary note on the method of Least Median of Squares reconsidered Trans Twelfth Prague Conf on Inform Theory, Statist Dec Functions and Rand Proc, Academy of Sciences of the Czech Republic, pp 254–259
Víšek JÁ (1996) Sensitivity analysis of M-estimates. Ann Inst Stat Math 48:469–495
Víšek JÁ (1998) Robust specification test. In: Marie Hušková, Petr Lachout & , Jan Ámos Víšek (eds) Proceedings of Prague Stochastics’98. Published by Union of Czechoslovak Mathematicians and Physicists, 1998, pp 581–586
Víšek J Á (2000) Regression with high breakdown point. In: Antoch J &, Dohnal G (eds) Proceeding of the conference ROBUST 2000. Published by The Union of the Czech Mathematicians and Physicists 2001, pp 324–356
Víšek JÁ (2002) Sensitivity analysis of M-estimates of nonlinear regression model: Influence of data subsets. Ann Inst Stat Math 54:261–290
Víšek JÁ (2006) Instrumental Weighted Variables - algorithm. In: Rizzi A, Vichi M (eds) Proceedings of COMPSTAT 2006. Physica-Verlag (Springer Company) Heidelberg 2006, pp 777–786
Víšek JÁ (2006) The least trimmed squares. Sensitivity study. In: Hušková M. &, Janžura M (eds) Proc Prague Stochastics 2006. matfyzpress, pp 728–738
Víšek JÁ (2010) Weak \(\sqrt {n}\)-consistency of the least weighted squares under heteroscedasticity. Acta Universit Carolinae, Math 2/51:71–82
Víšek JÁ (2010) Robust error-term-scale estimate. IMS Collections. Nonparametrics and Robustness in Modern Statistical Inference and Time Series Analysis: Festschrift for Jana Jurečková, vol 7 (2010), pp 254–267
Víšek JÁ (2011) Consistency of the least weighted squares under heteroscedasticity. Kybernetika 47:179–206
Víšek JÁ (2011) Empirical distribution function under heteroscedasticity. Stat 45:497–508
Víšek JÁ (2011) Robust estimation under broken orthogonality condition. The lecture given in the Nuffield College, Oxford, November 25, 2011 (the paper is still in a form of preprint, presentation is available on http://samba.fsv.cuni.cz/visek/Oxford/)
Víšek JÁ (2013) Least weighted squares with constraints and under heteroscedasticity. Bulletin Czech Econ Soc 20(31):21–54
Víšek JÁ (2014) Asymptotic representation of the instrumental weighted variables - theory and practice. Part II - numerical study. Bulletin Czech Econ Soc 21 (32):48–72
Víšek JÁ (2014) Instrumental weighted variables under heteroscedasticity. Part II. Numerical study. Submitted to Kybernetika
Wallace TD, Hussain A (1969) The Use of error components models in combining cross section with time series data. Econometrica 37:55–72
White H (1980) A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroscedasticity. Econometrica 48:817–838
Wooldridge JM (2006) Introductory Econometrics. A Modern Approach MIT Press Cambridge Massachusetts, second edition 2009
Zellner A (1962) An Efficient Method of Estimating Seemingly Unrelated Regressions and Tests for AggregationBias. J Am Stat Assoc 57:348–68
Zellner A (1965) Joint Estimation of relationships involving discrete random variables. Econ 33:382–394
Author information
Authors and Affiliations
Corresponding author
Additional information
The paper was written with the support of the Czech Science Foundation project P402/12/G097 DYME - Dynamic Models in Economics.
Rights and permissions
About this article
Cite this article
Víšek, J.Á. Estimating the Model with Fixed and Random Effects by a Robust Method. Methodol Comput Appl Probab 17, 999–1014 (2015). https://doi.org/10.1007/s11009-014-9432-5
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11009-014-9432-5