|
Abstract:
|
The absence of user-friendly software is still a major obstacle to the routine application of Bayesian methods. WinBUGS is the most widely used, but many will not bother to learn the language and its nuances unless they are already highly motivated to try Bayesian methods. I describe a GUI that facilitates the specification of a wide class of GLMMs for analysis using WinBUGS. The program, BugsXLA, not only allows the user to specify a model as one would in a package such as SAS or S+, but also aids the specification of priors and control of the MCMC run itself. Results are reported in a format that is intuitive to comprehend; additional predictions, contrasts as well as model checking summaries can also be easily obtained. It is hoped that by providing an easy way to develop WinBUGS code more statisticians and scientists working in industry will consider the Bayesian approach.
|
