Flexible estimation of serial correlation in nonlinear mixed models |
| |
Authors: | Jan Serroyen Marc Aerts Ellen Vloeberghs Peter Paul De Deyn Geert Verbeke |
| |
Institution: | 1. Department of Methodology and Statistics , University Maastricht , Peter Debyeplein 1, 6229 HA , Maastricht , The Netherlands;2. I-BioStat, Hasselt University , Agoralaan 1, 3590 , Diepenbeek , Belgium;3. Laboratory of Neurochemistry and Behaviour , University of Antwerp , Universiteitsplein 1, 2170 , Wilrijk , Belgium;4. I-BioStat, Hasselt University , Agoralaan 1, 3590 , Diepenbeek , Belgium;5. I-BioStat, Katholieke Universiteit Leuven , Kapucijnenvoer 35, 3000 , Leuven , Belgium |
| |
Abstract: | In the conventional linear mixed-effects model, four structures can be distinguished: fixed effects, random effects, measurement error and serial correlation. The latter captures the phenomenon that the correlation structure within a subject depends on the time lag between two measurements. While the general linear mixed model is rather flexible, the need has arisen to further increase flexibility. In addition to work done in the area, we propose the use of spline-based modeling of the serial correlation function, so as to allow for additional flexibility. This approach is applied to data from a pre-clinical experiment in dementia which studied the eating and drinking behavior in mice. |
| |
Keywords: | Alzheimer's disease dementia ordinary least squares random effect |
|
|