11-30

Nonlinear Mixed Effects Models: Approximations of the Fisher Information and Design

by Mielke, T.

 

Preprint series: 11-30, Preprints

MSC:
62K05 Optimal designs
62J02 General nonlinear regression

 

Abstract: The problem of the missing closed form representation of the probability density of the observations in nonlinear mixed effects models carries forward to the calculation of the Fisher information. Linearizations of the response function are often applied for approximating the underlying statistical model. The impact of different linearizations on the design of experiments will be briefly discussed in this article and an alternative motivation for an approximation of the Fisher information will be presented. The different results will be illustrated in the example of a simple population pharmacokinetic model.

Keywords: Fisher information, mixed effects models, nonlinear models, optimal design


The author(s) agree, that this abstract may be stored asfull text and distributed as such by abstracting services.

Letzte Änderung: 01.03.2018 - Ansprechpartner: Webmaster