In this paper we introduce a multiple testing procedure to assess a common covariance structure between k groups. The new test allows for a choice among eight different patterns arising from the three-term eigen decomposition of the group covariances. It is based on the closed testing principle and adopts local likelihood ratio tests. The approach reveals richer information about the underlying data structure than classical methods, the most common one being only based on homo/heteroscedasticity. At the same time, it provides a more parsimonious parameterization, whenever the constrained model is suitable to describe the real data. The new inferential methodology is then applied to some well-known data sets chosen from the multivariate literature. Finally, simulation results are presented to investigate its performance in different situations representing gradual departures from homoscedasticity and to evaluate the reliability of using the asymptotic χ2 to approximate the actual distribution of the local likelihood ratio test statistics.

Greselin, F., Punzo, A. (2013). Closed Likelihood Ratio Testing Procedures to Assess Similarity of Covariance Matrices. THE AMERICAN STATISTICIAN, 67(3), 117-128 [10.1080/00031305.2013.791643].

Closed Likelihood Ratio Testing Procedures to Assess Similarity of Covariance Matrices

GRESELIN, FRANCESCA;
2013

Abstract

In this paper we introduce a multiple testing procedure to assess a common covariance structure between k groups. The new test allows for a choice among eight different patterns arising from the three-term eigen decomposition of the group covariances. It is based on the closed testing principle and adopts local likelihood ratio tests. The approach reveals richer information about the underlying data structure than classical methods, the most common one being only based on homo/heteroscedasticity. At the same time, it provides a more parsimonious parameterization, whenever the constrained model is suitable to describe the real data. The new inferential methodology is then applied to some well-known data sets chosen from the multivariate literature. Finally, simulation results are presented to investigate its performance in different situations representing gradual departures from homoscedasticity and to evaluate the reliability of using the asymptotic χ2 to approximate the actual distribution of the local likelihood ratio test statistics.
Articolo in rivista - Articolo scientifico
Closed Testing Procedures, Common Principal Components, Eigen Decomposition, Homoscedasticity,Likelihood Ratio Tests, Proportional Covariance Matrices
English
2013
67
3
117
128
open
Greselin, F., Punzo, A. (2013). Closed Likelihood Ratio Testing Procedures to Assess Similarity of Covariance Matrices. THE AMERICAN STATISTICIAN, 67(3), 117-128 [10.1080/00031305.2013.791643].
File in questo prodotto:
File Dimensione Formato  
Greselin-2013-American Statistician-VoR.pdf

accesso aperto

Descrizione: Article
Tipologia di allegato: Author’s Accepted Manuscript, AAM (Post-print)
Licenza: Creative Commons
Dimensione 510.39 kB
Formato Adobe PDF
510.39 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/44359
Citazioni
  • Scopus 19
  • ???jsp.display-item.citation.isi??? 18
Social impact