## Abstract

In the present paper it will be argued that if a parameter value assigns zero probability to an open set containing the actual response, then that parameter value should be excluded from the parameter space. When this is done, the model becomes a restricted model, and sampling theory inferences should be focused on the sampling distribution of an hypothetical independent response from the restricted model.

The above situation may arise when probability densities vanish outside a compact set. This phenomenon arises frequently in the real world, but it is usually ignored for reasons of mathematical convenience. Although many statistical procedures remain substantially the same when we consider this restricted model, admissibility propertics may be drastically changed, and inferences which are known to be inadmissible may turn out to be really admissible. Thus, Stein's phenomenon concerning the inadmissibility of the sample mean as an estimator of the population mean of a*p*-variate normal distribution when*p*≥3may be explained by the fact that the distribution has a compact support but this has been ignored by reasons of mathematical convenience.

The introduction of a restricted model is also important in the study of coherence. Thus, it will be shown that Brunk's theory of countably additive coherence, which admits the use of countably additive improper priors, can be improved with the introduction of a restricted model. Thus, the theory will be unified because it will be proved that the posterior is really coherent if and only if it is a Bayes posterior, and it will be simplified because it will not be required that the prior be minimally compatible with the model.

## Key Words

Improper priors, real admissibility real coherence restricted model Stein's phenomenon, truncated spaces## AMS subject classification

62C15 62F15## Preview

Unable to display preview. Download preview PDF.

## References

- Armstrong, T.E. and W. Sudderth (1989). Locally coherent rates of exchange.
*Annals of Statistics*,**17**, 1394–1408.zbMATHMathSciNetGoogle Scholar - Bauer, H. (1981).
*Probability Theory and Elements of Measure Theory*. Academic Press, New York.zbMATHGoogle Scholar - Berger, J.O. (1985).
*Statistical Decision Theory and Bayesian Analysis*. Springer-Verlag, New York.zbMATHGoogle Scholar - Billingsley, P. (1968).
*Convergence in Probability Measures*. John Wiley, New York.Google Scholar - Brink, H.D. (1991). Fully coherent inference.
*Annals of Statistics*,**19**, 830–849.MathSciNetGoogle Scholar - Buchler, R.J. (1976). Coherent preferences.
*Annals of Statistics*,**4**, 1051–1064.MathSciNetGoogle Scholar - Buchler, R.J. (1971). Measuring information and uncertainty. In
*Foundations of Statistical Inference*(V.P. Godambe and D.A. Sproot eds.) Holt, Rinchart and Winston of Canada.Google Scholar - Dawid, A.P. and M. Stonc (1972). Expectation consistency of inverse probability distributions.
*Biometrika*,**59**, 486–489.zbMATHCrossRefMathSciNetGoogle Scholar - Dawid, A.P. and M. Stone (1973). Expectation consistency and generalized Bayes inference.
*Annals of Statistics*,**1**, 478–485.zbMATHMathSciNetGoogle Scholar - Freedman, D. and R. Purves (1969). Bayes' method for bookies.
*Annals of Mathematical Statistics*,**40**, 1177–1186.MathSciNetGoogle Scholar - Heath, D.C. and W.D. Sudderth (1978). O finitely additive priors.
*Annals of Statistics*,**6**, 333–345.zbMATHMathSciNetGoogle Scholar - Heath, D.C. and W.D. Sudderth (1989). Coherent inference from improper priors and from finitely additive priors.
*Annals of Statistics*,**17**, 907–919.zbMATHMathSciNetGoogle Scholar - Kass, R. (1980). The Riemannian structures of model spaces: A geometric approach to inference. Ph.D. Thesis, University of Chicago, Department of Statistics.Google Scholar
- Kass, R. (1989). The geometry of asymptotic inference.
*Statistical Science*,**4**, 188–219.zbMATHMathSciNetGoogle Scholar - Lane, D.A. (1981). Coherence and prediction.
*Bulletin of the International Statistical Institute*,**49**, 81–96.zbMATHMathSciNetGoogle Scholar - Lanc, D.A. and W.D. Sudderth (1983). Coherent and continuous inferences.
*Annals of Statistics*,**11**, 114–120.MathSciNetGoogle Scholar - Lanc, D.A. and W.D. Sudderth (1984). Coherent predictive inference.
*Sankhya, A*,**46**, 166–185.Google Scholar - Plante, A. (1991). An inclusion-consistent solution to the problem of absurd confidence intervals: I. Consistent exact confidence-interval estimation.
*The Canadian Journal of Statistics*,**19**, 389–398.zbMATHMathSciNetGoogle Scholar - Press, S.J. (1982).
*Applied Multivariate Analysis*, Krieger, Malabar, Florida.zbMATHGoogle Scholar - Regazzini, E. (1987). De Finetti's coherence and statistical inference.
*Annals of Statistics*,**15**, 845–864.zbMATHMathSciNetGoogle Scholar - Swartz, T. and C. Villegas (1992). Posterior probability and conditional coverage.
*Statistics and Probability Letters*,**14**, 169–173.zbMATHCrossRefMathSciNetGoogle Scholar - Skala, H.J. (1988). On σ-additive priors, σ-coherence, and the existence of posteriors. In
*Risk, Decision and Rationality*(B.R. Munier, ed.) Reidel, Dordrecht, 563–574.Google Scholar - Stein, C. (1956). Inadmissibility for the usual estimator of the mean of a multivariate normal distribution. Proceedings of the Third Berkeley Symposium of Mathematical Statistics,
**9**, 768–776.Google Scholar - Taylor, A.E. and D.C. Lay (1980).
*Introduction to Functional Analysis*, John Wiley, New York.zbMATHGoogle Scholar - Villegas, C. (1981). Inner statistical inference II.
*Annals of Statistics*,**9**, 768–776.zbMATHMathSciNetGoogle Scholar - Villegas, C. (1990). Bayesian inference in models with Euclidean structures.
*Journal of the American Statistical Association*,**85**, 1159–1164.zbMATHCrossRefMathSciNetGoogle Scholar - Wald, A. (1950).
*Statistical Decision Functions*, Chelsea, New York.zbMATHGoogle Scholar - Zacks, S. (1971).
*The Theory of Statistical Inference*. John Wiley, New York.Google Scholar