Bootstrap tests and confidence regions for functions of a covariance matrix. (English) Zbl 0607.62048

Beran, Rudolf; Srivastava, Muni S.

Ann. Stat. 13, 95-115 (1985).

Authors’ abstract: Bootstrap tests and confidence regions for functions of the population covariance matrix have the desired asymptotic levels provided model restrictions, such as multiple eigenvalues in the covariance matrix, are taken into account in designing the bootstrap algorithm.

Reviewer: E.Khmaladze

MSC:

62G10	Nonparametric hypothesis testing
62G15	Nonparametric tolerance and confidence regions
62H25	Factor analysis and principal components; correspondence analysis
62H20	Measures of association (correlation, canonical correlation, etc.)

Keywords:

principal component analysis; eigenvectors; bootstrap confidence regions; bootstrap critical values; normal model likelihood ratio; Bootstrap tests; population covariance matrix; eigenvalues