Abstract
The first order local influence approach is adopted in this paper to assess the local influence of observations to canonical correlation coefficients, canonical vectors and several relevant test statistics in canonical correlation analysis. This approach can detect different aspects of influence due to different perturbation schemes. In this paper, we consider two different kinds, namely, the additive perturbation scheme and the case-weights perturbation scheme. It is found that, under the additive perturbation scheme, the influence analysis of any canonical correlation coefficient can be simplified to just observing two predicted residuals. To do the influence analysis for canonical vectors, a scale invariant norm is proposed. Furthermore, by choosing proper perturbation scales on different variables, we can compare the different influential effects of perturbations on different variables under the additive perturbation scheme. An example is presented to illustrate the effectiveness of the first order local influence approach.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Cook, R. D. (1986). Assessment of local influence (with discussion), J. Roy. Statist. Soc. Ser. B, 48, 133–169.
Cook, R. D. and Weisberg, S. (1982). Residuals and Influence in Regression, Chapman and Hall, New York.
Fung, W. K. and Kwan, C. W. (1997). A note on local influence based on the normal curvature, J. Roy. Statist. Soc. Ser. B, 59, 839–843.
Jolliffe, I. T. and Lukudu, S. G. (1993). The influence of a single observation on some standard test statistics, Journal of Applied Statistics, 20, 143–151.
Lawrance, A. J. (1991). Local and deletion influence, Direction in Robust Statistics and Diagnostics, Part I, 141–157.
Lee, A. H. and Zhao, Y. (1996). Sensitivity of Pearson's goodness-of-fit statistic in generalized linear models, Comm. Statist. Theory Methods, 25, 143–157.
Lee, A. H. and Zhao, Y. (1997). Assessing influence on the goodness-of-link in generalized linear models, Statist. Probab. Lett. (to appear).
Muirhead, R. M. (1982). Aspects of Multivariate Statistical Theory, Wiley, New York.
Radhakrishnan, R. and Kshirsagar, A. M. (1981). Influence functions for certain parameters in multivariate analysis, Comm. Statist. Theory Methods, 10, 515–529.
Romanazzi, M. (1992). Influence in canonical correlation analysis, Psychometrika, 57, 237–259.
SAS/STAT User's Guide (1990). Version 6, Vol. 1, SAS Institute Inc., Cary, North Carolina.
Tanaka, Y. (1994). Recent advance in sensitivity analysis in multivariate statistical methods, J. Japanese Soc. Comput. Statist., 7, 1–25.
Wu, X. and Luo, Z. (1993). Second-order approach to local influence, J. Roy. Statist. Soc. Ser. B, 55, 929–936.
Author information
Authors and Affiliations
About this article
Cite this article
Gu, H., Fung, W.K. Assessing Local Influence in Canonical Correlation Analysis. Annals of the Institute of Statistical Mathematics 50, 755–772 (1998). https://doi.org/10.1023/A:1003717031299
Issue Date:
DOI: https://doi.org/10.1023/A:1003717031299