Statistical Methods and Applications

, Volume 16, Issue 1, pp 51–67 | Cite as

Robustness of Parameter Estimation Procedures in Multilevel Models When Random Effects are MEP Distributed

Original Article

Abstract

In this paper we examine maximum likelihood estimation procedures in multilevel models for two level nesting structures. Usually, for fixed effects and variance components estimation, level-one error terms and random effects are assumed to be normally distributed. Nevertheless, in some circumstances this assumption might not be realistic, especially as concerns random effects. Thus we assume for random effects the family of multivariate exponential power distributions (MEP); subsequently, by means of Monte Carlo simulation procedures, we study robustness of maximum likelihood estimators under normal assumption when, actually, random effects are MEP distributed.

Keywords

Hierarchical data ML and REML estimation Multivariate exponential power distribution 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Fang K-T, Kotz S, Ng K-W (1990) Symmetric multivariate and related distributions. Monograph on statistics and applied probability, vol 39. Chapman and Hall, LondonGoogle Scholar
  2. Ferrari PA, Solaro N (2002) Una proposta per le componenti erratiche del modello multilivello. In: Studi in onore di Angelo Zanella. Vita e Pensiero, Milano, pp 273–291Google Scholar
  3. Gómez E, Gómez-Villegas MA, Marin JM (1998) A multivariate generalization of the power exponential family of distributions. Commun Stat Theory Meth 27:589–600MATHGoogle Scholar
  4. Gori E, Vittadini G (1999) La valutazione dell’efficienza ed efficacia dei servizi alla persona. Impostazione e metodi. In: Qualità e valutazione nei servizi di pubblica utilità. ETAS, pp 121–241Google Scholar
  5. Hollander M, Wolfe DA (1973) Nonparametric statistical methods. Wiley, New YorkMATHGoogle Scholar
  6. Johnson ME (1987) Multivariate statistical simulation. Wiley, New YorkMATHGoogle Scholar
  7. Kreft IGG (1996) Are multilevel techniques necessary? An overview, including simulation studies. Technical report: http://www.calstatela.edu/ faculty/ikreftGoogle Scholar
  8. Laird NM, Ware JH (1982) Random-effects models for longitudinal data. Biometrics 38:963–974MATHCrossRefGoogle Scholar
  9. Lindsey JK (1999) Models for repeated measurements. Oxford University Press, OxfordGoogle Scholar
  10. Maas CJM, Hox JJ (2004) The influence of violations of assumptions on multilevel parameter estimates and their standard errors. Comput Stat Data Anal 46:427–440CrossRefMathSciNetGoogle Scholar
  11. Mardia KV (1970) Measures of multivariate skewness and kurtosis with applications. Biometrika 57:519–530MATHCrossRefMathSciNetGoogle Scholar
  12. Pinheiro JC, Bates MB (2000) Mixed-effects models in S and S-Plus. Springer, Berlin Heidelberg New YorkMATHGoogle Scholar
  13. Snijders TAB, Bosker RJ (1999) Multilevel analysis – an introduction to basic and advanced multilevel modeling. SAGE Publications, LondonMATHGoogle Scholar
  14. Solaro N (2004) Random variate generation from multivariate exponential power distribution. Statistica Applicazioni II, 2:25–44Google Scholar
  15. Verbeke G, Lesaffre E (1997) The effect of misspecifying the random-effects distribution in linear mixed models for longitudinal data. Comput Stat Data Anal 23:541–556MATHCrossRefMathSciNetGoogle Scholar
  16. Verbeke G, Molenberghs G (2000) Linear mixed models for longitudinal data. Springer, Berlin Heidelberg New YorkMATHGoogle Scholar

Copyright information

© Springer-Verlag 2006

Authors and Affiliations

  1. 1.Dipartimento di StatisticaUniversità di Milano-BicoccaMilanoItaly
  2. 2.Dipartimento di Scienze Economiche, Aziendali e StatisticheUniversità di MilanoMilanoItaly

Personalised recommendations