References of "Global Journal of Mathematical Sciences"
     in
Bookmark and Share    
Full Text
Peer Reviewed
See detailMinimal error rate of linear, quadratic and logistic rules in discriminant analysis
Glele Kakaï, R.; Palm, Rodolphe ULg

in Global Journal of Mathematical Sciences (2005), 4(1 & 2), 89-93

A simulation study has been used to evaluate the minimal error rate of three affectation rules and to compare five estimators of this error rate for two groups of observations, in 480 situations ... [more ▼]

A simulation study has been used to evaluate the minimal error rate of three affectation rules and to compare five estimators of this error rate for two groups of observations, in 480 situations characterized by the distribution and to the overlap of the populations, the number of variables, the sample size and the heteroscedasticity degree of the population under study. The results of this study suggest that the quadratic rule might be the best for heteroscedastic normal models. The linear rule showed better performance for homoscedastic normal or moderate non-normal models. The logistic rule is the best for severe non-normal models except when homoscedasticity occurs. As far as the comparison of five estimators is concerned, the results of the study indicate that eDS and eB are the best estimators of the minimal error rate for the linear rule, e5 for the quadratic rule and eD for the logistic rule. [less ▲]

Detailed reference viewed: 6 (0 ULg)
Full Text
Peer Reviewed
See detailA stepwise selection technique of the most discriminant parameters of two groups applied to isoberlinia stands in Benin.
Glele Kakaï, R.; Sinsin, Brice; Palm, Rodolphe ULg

in Global Journal of Mathematical Sciences (2005), 4(1 & 2), 107-111

We describe for two groups a stepwise selection method of the most discriminant parameters, based on the logistic rule and the e632 error rate estimator in discriminant analysis. This stepwise ... [more ▼]

We describe for two groups a stepwise selection method of the most discriminant parameters, based on the logistic rule and the e632 error rate estimator in discriminant analysis. This stepwise discriminant method can be used as an alternative method in non-normality or heteroscedasticity situations, where the stepwise linear discriminant analysis is not suitable. The implementation of this stepwise method on Isoberlinia woodland and tree savannah identifies the most discriminant parameters as the global dominant height, the dominant height of Isoberlinia trees and the number of species per hectare. For these parameters, the value of the error rate associated with the logistic rule and estimated by th e632 estimators is 24,6 %. [less ▲]

Detailed reference viewed: 18 (2 ULg)