TY - JOUR
AB - 214 Discussion on the Paper by Dr Dawid, Professor Stone and Dr Zidek [No.2, Professor D. J. BARTHOLOMEW (London School of Economics): The authors are to be congratulated not only on the content of the paper but also on the clarity of their exposition. Instead of burying their main result in mathematical abstraction they have taken the trouble to initiate the reader into the mysteries of the paradox by a succession of examples of real statistical interest. The ordinary reader is thus able to share the deepening perplexity of B and B as consistency, the fundamental tenet of the Bayesian creed, 1 2 comes under increasing strain. The danger of translating the higher mysteries of inference into the common tongue is obvious. Every humble practising statistician will want to take part in the debate and we must not be surprised if, when he does, he dismisses the whole exercise as unimportant. It seems clear that B and B are likely to come to much the same conclusions-especially 1 2 when they have plenty of data. But that is not the issue which is at stake. The real question, it seems to me, is "can Bayesian inference be made objective?" 
TI - Discussion on the Paper by Dr Dawid, Professor Stone and Dr Zidek
JF - Journal of the Royal Statistical Society Series B (Statistical Methodology)
DO - 10.1111/j.2517-6161.1973.tb00953.x
DA - 1973-01-01
UR - https://www.deepdyve.com/lp/oxford-university-press/discussion-on-the-paper-by-dr-dawid-professor-stone-and-dr-zidek-J02jzjMfsL
SP - 214
EP - 233
VL - 35
IS - 2
DP - DeepDyve
ER -