For each of the 9 sets of parameters (left column), 10000 samples of size 100 of the were simulated, F being the Log-normal distribution adjusted on Kelly and Rahn's data. The 95% confidence interval
A Bayesian framework is attractive in the context of prediction, but a fast recursive update of the predictive distribution has apparently been out of reach, in part because Monte Carlo methods are ge
Estimated relative L2 distances for each of the methods of combining subposterior samples to estimate posterior densities given the full data set. Results are included for the Bayesian logistic regres
