[Statlist] Sémin. STAT jeudi 18.05.2017- Daniel Flores Agreda

[ISTAT Messagerie]

SEMINAIRE DE STATISTIQUE

Institut de Statistique, Université de Neuchâtel, 2000 Neuchâtel - http://www2.unine.ch/statistics         

Jeudi 18 mai 2017,  11H00,  salle C14, 1er étage, Bâtiment G (bâtiment de chimie), Faculté des sciences, Av. de Bellevaux 51 

Daniel Flores Agreda, 

Research Center for Statistics - Geneva School of Economics and Management University of Geneva

Title: 
Bootstrap Inference in Generalized Linear Mixed Models Joint with Eva Cantoni

Abstract:
This talk aims to contribute to the discussion on the inference of Generalized Linear Mixed Models (GLMM). 
On a first stage, we review bootstrap schemes in GLMM and, inspired in part by schemes based on the random weighting of the Estimating Equations of a Gaussian Linear Mixed Model (LMM), we formulate a strategy based on (i) random weighting of the contributions to the Joint Likelihood of outcomes and random effects and (ii) the Laplace method for the approximation of integrals. 
On a second stage, we focus on the measure of uncertainty for predictions for random effects , customarily carried out via approximations to both Frequentist measures such as the Mean Squared Error of Prediction (MSEP) and others inherited from the Bayesian paradigm, such as the Conditional Variances (CV), and propose an implementation of our bootstrap method as an alternative way of computing these measures.