PREDICTIVE POSTERIOR DISTRIBUTIONS FROM A BAYESIAN VERSION OF A SLASHPINE YIELD MODEL

We formulate a traditional slash pine diameter distribution yield mode l in a Bayesian framework. We attempt to introduce as few new assumpti ons as possible. We generate predictive posterior samples for a number of stand Variables using the Gibbs sampler. The means of the samples agree well with the predictions from the published model. In addition, our model delivers distributions of outcomes, from which it is easy t o establish measures of uncertainty, e.g., Bayesian credible regions.