Use of the Probability Integral Transformation to Fit Nonlinear Mixed-Effects Models With Nonnormal Random Effects

摘要: This article describes a simple computational method for obtaining the maximum likelihood estimates (MLE) in nonlinear mixed-effects models when random effects are assumed to have nonnormal distribution. Many computer programs fitting models, such as PROC NLMIXED SAS, require that normal However, there is often interest either with or assessing sensitivity of inferences departures from normality assumption effects. When distribution, we show how probability integral transform can be used, conjunction standard statistical software (e.g., SAS), obtain MLEs. Specifically, used effect effect. The illustrated using gamma frailty model cl...