A default conjugate prior for variance components in generalized linear mixed models (Comment on article by Browne and Draper). (English) Zbl 1331.62148

Summary: For a scalar random-effect variance, W. J. Browne and D. Draper [ibid. 1, No. 3, 473–514 (2006; Zbl 1331.62125)] have found that the uniform prior works well. It would be valuable to know more about the vector case, in which a second-stage prior on the random effects variance matrix \(D\) is needed. We suggest consideration of an inverse Wishart prior for \(D\) where the scale matrix is determined from the first-stage variance.


62F15 Bayesian inference
62F40 Bootstrap, jackknife and other resampling methods
62D05 Sampling theory, sample surveys
62J12 Generalized linear models (logistic models)


Zbl 1331.62125
Full Text: DOI Euclid