All Times EDT
Keywords: software, R package, generalized linear mixed models, maximum likelihood, likelihood-based inference, mixed models, likelihood approximation, Monte Carlo
The R package glmm enables weighted likelihood-based inference for generalized linear mixed models with a canonical link by approximating the entire likelihood function and two derivatives. The model fitting function of this package, also called glmm, calculates and maximizes the Monte Carlo likelihood approximation to find Monte Carlo maximum likelihood estimates for the fixed effects and variance components. Additionally, the value, gradient vector, and Hessian matrix of the likelihood approximation are calculated at the Monte Carlo maximum likelihood estimates. Other functions in this package produce the variance-covariance matrix of the estimates, confidence intervals for the parameters, and Monte Carlo standard errors of the Monte Carlo maximum likelihood estimates. The package's ability to implement parallel computing can reduce users' computation time.