Statistical Inference under Symmetry

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.

[1]  J. Neumann.  Mathematische grundlagen der Quantenmechanik , 1935 .

[2]  J. VonNeumann.  Mathematische Grundlagen der Quantenmechanik , 1932 .

[3]  E. L. Lehmann,et al.  Theory of point estimation , 1950 .

[4]  C. Stein.  THE ADMISSIBILITY OF PITMAN'S ESTIMATOR OF A SINGLE LOCATION PARAMETER' , 1959 .

[5]  Donald Fraser,et al.  The fiducial method and invariance , 1961 .

[6]  L. Nachbin,et al.  The Haar integral , 1965 .

[7]  M. Stone,et al.  Right Haar Measure for Convergence in Probability to Quasi Posterior Distributions , 1965 .

[8]  Chester Hartman,et al.  Rejoinder by the Author , 1965 .

[9]  C. Stein.  Approximation of Improper Prior Measures by Prior Probability Measures , 1965 .

[10]  R. Buehler,et al.  Fiducial Theory and Invariant Estimation , 1966 .

[11]  D. Fraser.  The Structure of Inference. , 1969 .

[12]  James V. Bondar.  Structural Distributions Without Exact Transitivity , 1972 .

[13]  M. Stone,et al.  Marginalization Paradoxes in Bayesian and Structural Inference , 1973 .

[14]  James V. Bondar,et al.  Amenability: A survey for statistical applications of hunt-stein and related conditions on groups , 1981 .

[15]  D. Fraser,et al.  Inference and Linear Models. , 1981 .

[16]  R. A. Bailey.  A Unified Approach to Design of Experiments , 1981 .

[17]  A. P. Dawid,et al.  The Functional-Model Basis of Fiducial Inference , 1982 .

[18]  A. Dawid,et al.  Symmetry models and hypotheses for structured data layouts , 1988 .

[19]  J. Berger.  Statistical Decision Theory and Bayesian Analysis , 1988 .

[20]  M. L. Eaton.  Group invariance applications in statistics , 1989 .

[21]  T. Kariya.  Equivariant Estimation in a Model with an Ancillary Statistic , 1989 .

[22]  I. Helland.  Partial least squares regression and statistical models , 1990 .

[23]  R. A. Bailey,et al.  Strata for Randomized Experiments , 1991 .

[24]  I. Helland.  Simple Counterexamples against the Conditionality Principle , 1995 .

[25]  M. L. Eaton,et al.  The formal posterior of a standard flat prior in MANOVA is incoherent , 1995 .

[26]  L. Wasserman,et al.  The Selection of Prior Distributions by Formal Rules , 1996 .

[27]  Bradley Efron,et al.  R.A. Fisher In The 21St Century , 1997 .

[28]  P. McCullagh.  Linear Models, Vector Spaces, and Residual Likelihood , 1997 .

[29]  M. L. Eaton,et al.  A NEW PREDICTIVE DISTRIBUTION FOR NORMAL MULTIVARIATE LINEAR MODELS , 1998 .

[30]  Statistical inference under a fixed symmetry group , 1998 .

[31]  David R. Anderson,et al.  Model Selection and Multimodel Inference , 2003 .

[32]  Bradley Efron,et al.  R. A. Fisher in the 21st century (Invited paper presented at the 1996 R. A. Fisher Lecture) , 1998 .

[33]  I. Helland.  A Population Approach to Analysis of Variance Models , 1998 .

[34]  Restricted maximum likelihood from symmetry , 1999 .

[35]  Tore Schweder,et al.  Frequentist Analogues of Priors and Posteriors , 1999 .

[36]  M. L. Eaton,et al.  Consistency and strong inconsistency of group-invariant predictive inferences , 1999 .

[37]  I. Helland.  Reduction of regression models under symmetry , 2000 .

[38]  I. Helland.  Some theoretical aspects of partial least squares regression , 2001 .

[39]  Leo Breiman,et al.  Statistical Modeling: The Two Cultures (with comments and a rejoinder by the author) , 2001 .

[40]  S. James Press,et al.  Subjective and Objective Bayesian Statistics , 2002 .

[41]  William D. Sudderth,et al.  Group invariant inference and right Haar measure , 2002 .

[42]  David R. Anderson,et al.  Model selection and multimodel inference : a practical information-theoretic approach , 2003 .

[43]  I. Helland.  Quantum theory as a statistical theory under symmetry , 2003, quant-ph/0411174.

[44]  Stephen E. Fienberg,et al.  Testing Statistical Hypotheses , 2005 .

[45]  Extended statistical modelling under symmetry: The link towards quantum mechanics , 2005 .

[46]  A. P. Dawid,et al.  Invariant Prior Distributions , 2006 .