Standard Errors for EM Estimates in Generalized Linear Models with Random Effects
Authors: Friedl, Herwig1; Kauermann, Göran2
Source: Biometrics, Volume 56, Number 3, September 2000 , pp. 761-767(7)
Publisher: Blackwell Publishing
- Free Content
- New Content
- Subscribed Content
- Free Trial Content
Abstract:
Summary. A procedure is derived for computing standard errors of EM estimates in generalized linear models with random effects. Quadrature formulas are used to approximate the integrals in the EM algorithm, where two different approaches are pursued, i.e., Gauss-Hermite quadrature in the case of Gaussian random effects and nonparametric maximum likelihood estimation for an unspecified random effect distribution. An approximation of the expected Fisher information matrix is derived from an expansion of the EM estimating equations. This allows for inferential arguments based on EM estimates, as demonstrated by an example and simulations.Keywords: EM algorithm; Estimating equations; Gauss-Hermite quadrature; Mixture model; Nonparametric maximum likelihood estimation; Random effect model
Document Type: Research article
DOI: 10.1111/j.0006-341X.2000.00761.x
Affiliations: 1: Institute of Statistics, Technical University Graz, Steyrergasse 17, 8010 Graz, Austria, Email: friedl@stat.tu-graz.ac.at 2: Institute of Statistics, Ludwig-Maximilians-University Munich, Akademiestrasse 1, 80796 Munich, Germany, Email: kauerman@stat.uni-muenchen.de
- Free Content
- New Content
- Subscribed Content
- Free Trial Content
Click here for Page Help