Loading...
Loading...

Go to the content (press return)

Parameter estimation of Poisson generalized linear mixed models based on three different statistical principles: a Simulation Study

Author
Casals, M.; Langohr, K.; Carrasco, J.; Rönnegård, L.
Type of activity
Presentation of work at congresses
Name of edition
XIV Conferencia Española de Biometría
Date of publication
2013
Presentation's date
2013-05-24
Book of congress proceedings
XIV Conferencia Española de Biometría, Ciudad Real 22-24 Mayo 2013
First page
206
Last page
208
URL
http://areaestadistica.uclm.es/biometria2013/ResumenesXIVCEB.pdf Open in new window
Abstract
Generalized linear mixed models (GLMMs) are a flexible approach to fit non-normal data. The main difficulty of GLMMs is the parameter estimation because it is often not viable to obtain an analytic solution that allows maximizing the marginal likelihood of data. Hence, it is possible to find different principles to fit a GLMM implemented in the main statistical software packages. The purpose of this study is to compare the performance of these GLMM estimation methods via a simulation study with ...
Keywords
Estimation methods, GLMM, Statistical principles, overdispersion
Group of research
GRBIO - Biostatistics and Bioinformatics Research Group

Participants

  • Casals, Martí  (author and speaker )
  • Langohr, Klaus  (author and speaker )
  • Carrasco Jordan, Josep Lluís  (author and speaker )
  • Rönnegård, Lars  (author and speaker )