New influence measures in polytomous logistic regression models based on phi-divergence measures Articles
- May 2014
Digital Object Identifier (DOI)
International Standard Serial Number (ISSN)
Electronic International Standard Serial Number (EISSN)
- We consider the asymptotic distribution of divergence-based influence measures which are an extension for polytomous logistic regression of an influence measure proposed in Johnson (1985), for binary logistic regression. A numerical example compares the classical Cook's distance with the divergence based influence measures.
- divergence measures; cook's distance; polytomous regression model; primary 62h15; secondary 62h17