ABSTRACT
Most skills are best learned not through reading or quiet reflection, but through actual practice. This is certainly true of data analysis, an area with few hard-and-fast rules that is as much art as science. It is especially true of Bayes and empirical Bayes data analysis, with its substantial computing component and the challenging model options encouraged by the breadth of the methodology. The recent books edited by Gatsonis et al. (1993; 1995; 1997; 1999; 2002a,b), Gilks et al. (1996), and Berry and Stangl (1996; 2000) offer a broad range of challenging and interesting case studies in Bayesian data analysis and meta-analysis, most using modern MCMC computational methods. In addition, Bayesian Analysis, the official journal of the International Society for Bayesian Analysis (ISBA), publishes a great many applications of Bayesian methods to challenging datasets; see ba.stat.cmu.edu.
