Numerical consistent estimates in the multivariate linear mixed-effects model and application to the malaria infection study
schedule le lundi 12 novembre 2018 de 17h00 à 18h00
Organisé par : A. Lefebvre, N. Meyer, O. Safsafi, T. Touati
Intervenant : Eric Adjakossa (LPSM)
Lieu : Jussieu, salle Paul Lévy, couloir 16-26, salle 113
Sujet : Numerical consistent estimates in the multivariate linear mixed-effects model and application to the malaria infection study
In this work we provide both ML and REML criteria for consistently estimating multivariate linear mixed-effects models parameters. By factorizing the random effects covariance matrix, we provide an explicit expression of the profiled deviance through a reparameterization of the model. Beside its robustness regarding starting points, the approach enables an improved computational in comparison with the traditional approaches such as the EM algorithm. We provide an illustration by using the real-life data concerning a study of the child's immune against malaria in Africa.