This paper was presented at the opening session of the First Catalan International Symposium on Statistics, held in Barcelona, September 1983, in honour to the late Gwilym M. Jenkins. It is therefore a special paper in its format. It is a great honour for our Journal to publish it (Editor's note).
Scientific learning is seen as an iterative process employing Criticism and Estimation. Sampling theory use of predictive distributions for model criticism is examined and also the implications for significance tests and the theory of precise measurement. Normal theory examples and ridge estimates are considered. Predictive checking functions for transformation, serial correlation, and bad values are reviewed as is their relation with Bayesian options. Robustness is seen from a Bayesian view point...
Page 1