Skip to main content


Fig. 1 | BMC Medical Research Methodology

Fig. 1

From: Prediction models for clustered data with informative priors for the random effects: a simulation study

Fig. 1

The random effects distribution divided into multiple truncated areas of equal proportions in 3 different scales. A truncated area contains either half, one third, or one fifth of the distribution. Based on elicited expert knowledge for each cluster, a particular truncated area from each scale is chosen and used as prior distribution for the random effect of the cluster. We considered a prior distribution that contains 1/2 of the distribution as low informative, 1/3 as medium informative, and 1/5 as highly informative

Back to article page