Prediction in Multilevel Generalized Linear Models

Abstract

Summary. We discuss prediction of random effects and of expected responses in multilevel generalized linear models. Prediction of random effects is useful for instance in small area estimation and disease mapping, effectiveness studies and model diagnostics. Prediction of expected responses is useful for planning, model interpretation and diagnostics. For prediction of random effects, we concentrate on empirical Bayes prediction and discuss three different kinds of standard errors; the posterior standard deviation and the marginal prediction error standard deviation (comparative standard errors) and the marginal sampling standard deviation (diagnostic standard error). Analytical expressions are available only for linear models and are provided in an appendix. For other multilevel generalized linear models we present approximations and suggest using parametric bootstrapping to obtain standard errors. We also discuss prediction of expectations of responses or probabilities for a new unit in a hypothetical cluster, or in a new (randomly sampled) cluster or in an existing cluster. The methods are implemented in gllamm and illustrated by applying them to survey data on reading proficiency of children nested in schools. Simulations are used to assess the performance of various predictions and associated standard errors for logistic random-intercept models under a range of conditions.

Variations in outcome and costs among NHS providers for common surgical procedures: econometric analyses of routinely collected data

“…This can be interpreted as the estimated reliability of the fixed-effect estimate as a measure of γ j . 54 Alternatively, the random-effect estimate itself can be regarded as a precision-weighted estimate. 55 Based on estimates of γ j , we can now proceed to describe hospital performance.…”

“…This parameter is not directly estimated but can be evaluated in post-estimation using Bayes' theorem with variance estimates entered for the unknown population parameters, which is a technique known as Empirical Bayes prediction. 54 The empirical Bayes estimates (EBEs) are estimated in two distinct steps. First, the risk-adjustment model is estimated using individual patient-level data while recognising the clustering of patients within hospitals.…”

Estimating Modifying Effect of Age on Genetic and Environmental Variance Components in Twin Models

“…We found in the simulation study that the posterior mean b is often quite close to the mode b b; suggesting that the posterior distribution is highly symmetric. The conditional MSE of prediction VarðbjyÞ is proposed to be a better estimate of prediction error than Varðb 2 bÞ (Skrondal and Rabe-Hesketh 2009), which thus is adopted to compute the C.I. 's in our analyses.…”

Resisting the extortion racket: an empirical analysis

“…The integral required to calculate the log-likelihood is approximated by using the mean-variance adaptive Gauss-Hermite quadrature (Skrondal and Rabe-Hesketh, 2004) Prediction of random effects and expected responses is also often required. An extensive treatment of this topic is addressed by Skrondal and Rabe-Hesketh (2004;2009).…”