The current popularity of the use of Bayes approach to problems of applied statistics and statistical inference has cast the problem of "goodness of fit" and appropriatness of the assumed model into center stage under the full glare of experimenters' and statisticians' spotlights.nIn a recent paper, Anscombe (1963) approaches the problem of "goodness of fit" of observations with a hypothesis in the spirit of significance testing, A wonderful opening sentence of his paper reads: "To anyone sympathetic with the current neo-Bernouillian neo-Bayesian Ramseyeaque Finettist Savageous movement in statistics, the subject of testing goodness of fit is something of an embarrassment,"nThe purpose of this paper is to present a procedure for the following age-old goodness of fit problem, namely.
展开▼
