Influence Diagnostics in the Common Canonical Variates Model

作者: Hong Gu , Wing K. Fung

DOI: 10.1023/A:1017533528342

关键词:

摘要: As a generalization of the canonical correlation analysis to k random vectors, common variates model was recently proposed based on assumption that have same coefficients in all sets variables, and is applicable many cases. In this article, we apply local influence method study impact minor perturbations data. The non-standard because restrictions imposed coefficients. Besides investigating joint observations, also obtain elliptical norm empirical function as special case diagnostics. Based diagnostics, find results for female water striders data set largely affected by omitting just one single observation.

参考文章(13)
Bernard D. Flury, Beat E. Neuenschwander, Simultaneous diagonalization algorithms with applications in multivariate statistics Proceedings of the conference on Approximation and computation : a fetschrift in honor of Walter Gautschi: a fetschrift in honor of Walter Gautschi. pp. 179- 205 ,(1994) , 10.1007/978-1-4684-7415-2_12
Yao-Ting Zhang, Kai-Tang Fang, Generalized Multivariate Analysis ,(1990)
A.J. Lawrance, Local and deletion influence ,(1990)
J. R. KETTENRING, Canonical analysis of several sets of variables Biometrika. ,vol. 58, pp. 433- 451 ,(1971) , 10.1093/BIOMET/58.3.433
BEAT E. NEUENSCHWANDER, BERNARD D. FLURY, Common canonical variates Biometrika. ,vol. 82, pp. 553- 560 ,(1995) , 10.1093/BIOMET/82.3.553
Bernhard N. Flury, Common Principal Components in k Groups Journal of the American Statistical Association. ,vol. 79, pp. 892- 898 ,(1984) , 10.1080/01621459.1984.10477108
ROBERT SCHALL, TIMOTHY T. DUNNE, A note on the relationship between parameter collinearity and local influence Biometrika. ,vol. 79, pp. 399- 404 ,(1992) , 10.1093/BIOMET/79.2.399
N. Billor, R.M. Loynes, Local influence: a new approach Communications in Statistics-theory and Methods. ,vol. 22, pp. 1595- 1611 ,(1993) , 10.1080/03610929308831105