rpql offers fast joint selection of fixed and random effects in Generalized Linear Mixed Model (GLMMs) via regularization. Specifically the penalized quasi-likelihood (PQL, Breslow and Clayton, 1993) is used as a loss function, and penalties are added on to perform fixed and random effects selection e.g., the lasso (Tibshirani, 1996) penalty. This method of joint selection in GLMMs, referred to regularized PQL, is very fast compared to information criterion and hypothesis testing, and has attractive large sample properties (Hui et al., 2016). Its performance however may not be great if the amount of data to estimate each random effect is not large, i.e. the cluster size is not large.
tools:::Rd_package_author("rpql")
Maintainer: tools:::Rd_package_maintainer("rpql")
| Package: | rpql |
| Type: | Package |
| Version: | 0.8.1 |
| Date: | 2023-08-01 |
| License: | GPL-2 |
Breslow, N. E., and Clayton, D. G. (1993). Approximate inference in generalized linear mixed models. Journal of the American Statistical Association, 88, 9-25.
Hui, F.K.C., Mueller, S., and Welsh, A.H. (2017). Joint Selection in Mixed Models using Regularized PQL. Journal of the American Statistical Association, 112, 1323-1333.
Tibshirani, R. (1996). Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society. Series B (Methodological), 58, 267-288.
## Please see examples in help file for the rpql function
Run the code above in your browser using DataLab