Test

, Volume 1, Issue 1, pp 105–121

Estimative and predictive distances

  • Ann F. S. Mitchell
Article

Summary

Methods of estimating distances between members of (r, r) exponential families are considered. The first replaces the parameters in the geodesic distance associated with the information metric by their maximum likelihood estimates. The second is based on the family of predictive densities corresponding to Jeffrey’s invariant prior, using the sufficient statistics as co-ordinates of a Riemannian manifold. In all examples considered, the resulting estimative and predictive distances differ in form by only a simple multiple, the predictive distance being the shorter, and interesting geometrical relationships associated with flatness are also observed. Finally, the effect of the conjugate priors on distances and flatness is considered.

Keywords

Rao’s distance (r, r) exponential families Jeffreys’ invariant prior Predictive density Flatness Conjugate priors 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Aitchison, J. (1975). Goodness of prediction fit.Biometrika 62, 547–554.CrossRefMathSciNetMATHGoogle Scholar
  2. Aitchison, J. and Kay, J. W. (1975). Principles, practice and performance in decision-making in clinical medicine.Proc. 1973 NATO Conference on the Role and Effectiveness of Decision Theory. London: English Universities Press. pp. 252–273.Google Scholar
  3. Amari, S.-I. (1985).Differential—Geometrical Methods in Statistics, Lecture Notes in Statistics28. Berlin: Springer-Verlag.MATHGoogle Scholar
  4. Atkinson, C. and Mitchell, A. F. S. (1972). Rao’s distance measure. (UnpublishedTech. Rep.) Imperial College, U.K.Google Scholar
  5. Atkinson, C. and Mitchell, A. F. S. (1981). Rao’s distance measure.Sankyã A 43, 345–365.MathSciNetMATHGoogle Scholar
  6. Bernardo, J. M. (1979). Reference posterior distributions for Bayesian inference (with discussion).J. Roy. Statist. Soc. B 41, 113–147.MathSciNetMATHGoogle Scholar
  7. Critchley, F., Marriott, P. and Salmon, M. (1992). distances in statistics. Proc.XXXVI Riunione Scientifica, Societa Italiana di Statistica, Rome: CISU, pp. 39–60.Google Scholar
  8. Geisser, S. (1966). Predictive discrimination.Multivariate Analysis. (P. R. Krishnaiah, ed.) New York: Academic Press, pp. 149–163.Google Scholar
  9. Jeffreys, H. (1948)Theory of Probability, Oxford: University Press.MATHGoogle Scholar
  10. Kullback, S. and Leibler, R. A. (1951). On information and sufficiency.Ann. Math. Statist. 22, 525–540.MathSciNetGoogle Scholar
  11. Mahalanobis, P. C. (1936). On the generalized distance in statistics.Proc. Nat. Inst. Sci. India A 2, 49–55.MATHGoogle Scholar
  12. Mitchell, A. F. S. and Krzanowski, W. (1985). The Mahalanobis distance and elliptic distributions.Biometrika 72, 464–467.CrossRefMATHGoogle Scholar
  13. Mitchell, A. F. S. (1988). Statistical manifolds of univariate elliptic distributions.Internat. Statist. Rev. 56, 1–16.MathSciNetCrossRefMATHGoogle Scholar
  14. Mitchell, A. F. S. (1989). The information matrix, skewness tensor and α-connections for the general multivariate elliptic distribution.Ann. Inst. Statist. Math. 41, 289–304.CrossRefMathSciNetMATHGoogle Scholar
  15. Murray, G. D. (1977). A note on the estimation of probability density functions.Biometrika 64, 150–152.CrossRefMathSciNetMATHGoogle Scholar
  16. Rao, C. R. (1945). Information and the accuracy attainable in the estimation of statistical parameters.Bull. Calcutta Math. Soc. 37, 81–91.MathSciNetMATHGoogle Scholar
  17. Rao, C. R. (1949). On the distance between two populations.Sankyã 9, 246–248.Google Scholar

Copyright information

© SEIO 1992

Authors and Affiliations

  • Ann F. S. Mitchell
    • 1
  1. 1.Department of MathematicsImperial CollegeLondonUK

Personalised recommendations