International Statistical Review

Statistical Inference under Symmetry

Inge S. Helland

Full-text: Access denied (no subscription detected)

We're sorry, but we are unable to provide you with the full text of this article because we are not able to identify you as a subscriber. If you have a personal subscription to this journal, then please login. If you are already logged in, then you may need to update your profile to register your subscription. Read more about accessing full-text

Abstract

We explore the consequences of adjoining a symmetry group to a statistical model. Group actions are first induced on the sample space, and then on the parameter space. It is argued that the right invariant measure induced by the group on the parameter space is a natural non-informative prior for the parameters of the model. The permissible sub-parameters are introduced, i.e., the subparameters upon which group actions can be defined. Equivariant estimators are similarly defined. Orbits of the group are defined on the sample space and on the parameter space; in particular the group action is called transitive when there is only one orbit. Credibility sets and confidence sets are shown (under right invariant prior and assuming transitivity on the parameter space) to be equal when defined by permissible sub-parameters and constructed from equivariant estimators. The effect of different choices of transformation group is illustrated by examples, and properties of the orbits on the sample space and on the parameter space are discussed. It is argued that model reduction should be constrained to one or several orbits of the group. Using this and other natural criteria and concepts, among them concepts related to design of experiments under symmetry, leads to links towards chemometrical prediction methods and towards the foundation of quantum theory.

Article information

Source
Internat. Statist. Rev., Volume 72, Number 3 (2004), 409-422.

Dates
First available in Project Euclid: 8 December 2004

Permanent link to this document
https://projecteuclid.org/euclid.isr/1102516480

Keywords
Confidence sets Credibility sets Group Invariance Invariant measure Loss Non-informative prior Optimal estimator Objective Bayes Orbit Partial least squares regression Permissible sub-parameter Pitman estimator Quantum mechanics Right invariant prior Risk Symmetry Transitivity

Citation

Helland, Inge S. Statistical Inference under Symmetry. Internat. Statist. Rev. 72 (2004), no. 3, 409--422. https://projecteuclid.org/euclid.isr/1102516480


Export citation

References

  • [1] Bailey, R.A. (1981). A unified approach to design of experiments. J.R. Statist. Soc. A, 144, 214-223.
  • [2] Bailey, R.A. (1991). Strata for randomized experiments. J.R. Statist. Soc. B, 53, 27-78.
  • [3] Bailey, R.A. (2003). Design of Comparative Experiments. Book to appear. Draft available at\newline http://www.maths.qmw.ac.uk/ rab/DOEbook/
  • [4] Berger, J.O. (1985). Statistical Decision Theory and Bayesian Analysis. New York: Springer-Verlag.
  • [5] Bondar, J.V. (1972). Structural distributions without exact transitivity. Ann. Math. Statist., 43, 326-339.
  • [6] Bondar, J.V. & Milnes, P. (1981). Amenability: A survey for statistical applications of Hunt-Stein and related conditions on groups. Z. Wahrscheinlichkeitstheorie verw. Gebiete, 57, 103-128.
  • [7] Bondesson, L. (1982). Equivariant estimators. In Encyclopedia of Statistical Sciences, Eds. S. Kotz, N.L. Johnson and\linebreak C.B. Read. New York: John Wiley.
  • [8] Breiman, L. (2001). Statistical modelling: The two cultures. Statistical Science, 16, 199-231.
  • [9] Burnham, K.P. & Anderson, D.R. (2002). Model Selection and Multimodel Inference. A Practical Information-Theoretic Approach. New York: Springer-Verlag.
  • [10] Dawid, A.P. (1983). Invariant prior distributions. In Encyclopedia of Statistical Sciences, Eds. S. Kotz, N.L. Johnson and\linebreak C.B. Read. New York: John Wiley.
  • [11] Dawid, A.P. (1988). Symmetry models and hypotheses for structural data layouts. J.R. Statist. Soc. B, 50, 1-34.
  • [12] Dawid, A.P. & Stone, M. (1982). The functional-model basis of fiducial inference (with discussion). Ann. Statist., 10, 1054-1067.
  • [13] Dawid, A.P., Stone, M. & Zidek, J.V. (1973). Marginalization paradoxes in Bayesian and structural inference, J.R. Statist. Soc. B, 35, 189-233.
  • [14] Eaton, M.L. (1989). Group Invariance Applications in Statistics. Institute of Mathematical Statistics and American Statistical Association, Hayward, California.
  • [15] Eaton, M.L. & Sudderth, W.D. (1993). Prediction in a multivariate setting: Coherence and incoherence. Sankhya A, 55, 481-493.
  • [16] Eaton, M.L. & Sudderth, W.D. (1995). The formal posterior of a standard flat prior in MANOVA is incoherent. J. Italian Statist. Soc., 2, 251-270. Abstract can also be found in the ISI/STMA publication
  • [17] Eaton, M.L. & Sudderth, W.D. (1998). A new predictive distribution for normal multivariate linear models. Sankhya A, 60, 363-382. Abstract can also be found in the ISI/STMA publication
  • [18] Eaton, M.L. & Sudderth, W.D. (1999). Consistency and strong inconsistency of group-invariant predictive inferences. Bernoulli, 5, 833-854. Abstract can also be found in the ISI/STMA publication
  • [19] Eaton, M.L. & Sudderth, W.D. (2002). Group invariant inference and right Haar measures. J. Stat. Planning and Infer., 103, 87-99. Abstract can also be found in the ISI/STMA publication
  • [20] Efron, B. (1998). R.A. Fisher in the 21st century. Statistical Science, 13, 95-122.
  • [21] Fraser, D.A.S. (1961). The fiducial method and invariance. Biometrika, 48, 261-280.
  • [22] Fraser, D.A.S. (1968). The Structure of Inference. New York: Wiley.
  • [23] Fraser, D.A.S. (1979). Inference and Linear models. New York: McGraw-Hill.
  • [24] Helland, I.S. (1990). Partial least squares regression and statistical models. Scand. J. Statist., 17, 97-114. Abstract can also be found in the ISI/STMA publication
  • [25] Helland, I.S. (1995). Simple counterexamples against the conditionality principle. Amer. Statistician, 49, 351-356; discussion: 50, 382-386. Abstract can also be found in the ISI/STMA publication
  • [26] Helland, I.S. (1998). A population approach to analysis of variance models. Scand. J. Statistics, 25, 3-15. Abstract can also be found in the ISI/STMA publication
  • [27] Helland, I.S. (1999a). Restricted maximum likelihood from symmetry. Statistical Research Report No. 14/1999. Department of Mathematics, University of Oslo. Available at http://folk.uio.no/ingeh/publ.html.
  • [28] Helland, I.S. (2001a). Reduction of regression models under symmetry. Invited contribution to: M. Viana and D. Richards (Eds.) Algebraic Methods in Statistics. Contemporary Mathematics Series of the American Mathematical Society.
  • [29] Helland, I.S. (2001b). Some theoretical aspects of partial least squares regression. Chemometrics and Intelligent Laboratory Systems, 58, 97-107.
  • [30] Helland, I.S. (2002a). Statistical inference under a fixed symmetry group. Preprint, http://folk.uio.no/ingeh/publ.html.
  • [31] Helland, I.S. (2002b). Discussion of McCullagh, P. (2002) What is a statistical model? Ann. Statistics, 30, 1225-1310.
  • [32] Helland, I.S. (2003a). Extended statistical modelling under symmetry: The link towards quantum mechanics. Submitted. Available at http://folk.uio.no/ingeh/publ.html.
  • [33] Helland, I.S. (2003b). Quantum theory as a statistical theory under symmetry and complementarity. Submitted. Available at http://folk.uio.no/ingeh/publ.html.
  • [34] Hora, R.B. & Buehler, R.J. (1966). Fiducial theory and invariant estimation. Ann. Math. Statist., 37, 643-656.
  • [35] Kariya, T. (1989). Equivariant estimation in a model with an ancillary statistic. Ann. Statist., 17, 920-928. Abstract can also be found in the ISI/STMA publication
  • [36] Kass, R.E. & Wasserman, L. (1996). The selection of prior distributions by formal rules. J. Amer. Stat. Ass., 91, 1343-1370. Abstract can also be found in the ISI/STMA publication
  • [37] Lehmann, E.L.(1959). Testing Statistical Hypotheses. New York: Wiley.
  • [38] Lehmann, E. & Casella, G. (1998). Theory of Point Estimation. New York: Springer.
  • [39] McCullagh, P. (1996). Linear models, vector spaces, and residual likelihood. In Modelling Longitudinal and Spatially Correlated Data, Eds. Gregoire et al.. Springer Lecture Notes No. 122, pp. 1-10.
  • [40] Nachbin, L. (1965). The Haar Integral. Princeton, NJ: Van Nostrand.
  • [41] Press, S.J. (2003). Subjective and Objective Bayesian Statistics. Wiley.
  • [42] Schweder, T. & Hjort, N.L. (2003). Frequentist Analogues of Priors and Posteriors. In Econometrics and the Philosophy of Economics. Theory-Data Confrontations in Economics, Ed. B.P. Stigum. Princeton University Press.
  • [43] Stein, C. (1959). The admissibility of Pitman's estimator of a single location parameter. Ann. Math. Statist., 30, 970-979.
  • [44] Stein, C. (1965). Approximation of improper prior measures by prior probability measures. In Bernoulli, 1713; Bayes, 1763; Laplace, 1813, Eds. J. Neyman and L.M. LeCam, pp. 217-240. Berlin: Springer-Verlag.
  • [45] Stone, M. (1965). Right Haar measures of convergence in probability to invariant posterior distributions. Ann. Math. Statist., 36, 440-453.
  • [46] von Neumann, J. (1932). Matematische Grundlagen der Quantenmechanik. Berlin: Springer.
  • [47] Zacks, S. (1971). The Theory of Statistical Inference. New York: Wiley.