CMStatistics 2021: Start Registration
View Submission - CMStatistics
B0296
Title: Generalized functional linear models under non-random sampling Authors:  Christelle Judith Agonkoui - IMSP (Benin)
Feriel Bouhadjera - University of Montpellier (France)
Sophie Dabo - University of Lille (France) [presenting]
Abstract: A functional binary choice model is explored in a non-random sample design. In other words, a model is considered in which the response is binary, the explanatory variables are functional, and the sample is stratified with respect to the values of the response variable. A dimension reduction of the spaces of the explanatory random functions based on Karhunen-Loeve expansion is used to define a conditional maximum likelihood estimate of the model. Based on this formulation, several asymptotic properties are given. A finite sample study is proposed to compare the proposed method with the ordinary maximum likelihood method, which ignores the nature of the sampling. The potential of the functional model for integrating special non-random features of the sample, which would have been difficult to see otherwise, is outlined.