Influence functions for certain parameters in discriminant analysis
This paper derives the influence function for a parametric function that corresponds to the residual Wilks' \(\Lambda\) used in testing goodness of fit of a single hypothetical discriminant function in discriminant analysis with more than two groups. Influence functions for parameters that correspond to the direction and collinearity factors of the residual Wilks's \(\Lambda\) are also derived. These influence functions can be used to develop techniques for outlier detection in multivariate data when a single discriminant function is adequate while discriminating among several multivariate normal populations.
- Infiuence functions for certain parameters in discriminant analysis when a single discriminant function is not adequate
- scientific article; zbMATH DE number 801095
- INFLUENCE FUNCTION OF THE LIKELIHOOD RATIO TEST STATISTIC FOR MULTIVARIATE NORMAL SAMPLE
- Influence functions for linear discriminant analysis: sensitivity analysis and efficient influence diagnostics
- Influence in principal components analysis
- Detecting influential observations for estimated probabilities in multiple discriminant analysis
- The moment of inertia and the linear discriminant function
- Partial influence functions
- Influence functions for linear discriminant analysis: sensitivity analysis and efficient influence diagnostics
- Infiuence functions for certain parameters in discriminant analysis when a single discriminant function is not adequate
- INFLUENCE FUNCTION OF THE LIKELIHOOD RATIO TEST STATISTIC FOR MULTIVARIATE NORMAL SAMPLE
- scientific article; zbMATH DE number 1782272 (Why is no real title available?)
This page was built for publication: Influence functions for certain parameters in discriminant analysis
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1061430)
