No label defined (Q3313087)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 3842949
Language Label Description Also known as
default for all languages
No label defined
    English
    No label defined
    scientific article; zbMATH DE number 3842949

      Statements

      0 references
      1983
      0 references
      hypotheses tests
      0 references
      groups of linear transformations
      0 references
      general invariance concept
      0 references
      formal-algebraic approach
      0 references
      non-Gaussian distributions
      0 references
      If the distribution of a random vector X is invariant under a group of transformations, one can analyze the testing and distributional properties of an inference problem in greater detail than usual. This was first shown by the classical work of G. Hunt and C. Stein. Several types such as permutations, rotations, or translation invariances have been investigated in the literature. The author formulates a very general invariance concept which includes several previous works and announces four results. It may be described in his own words as follows: ''In this note we develop a general formal-algebraic approach to the solution of problems in this class (of distributions which are invariant under groups of transformations). The universality of the approach permits us to extend the main results to the case of non-Gaussian distributions having a symmetry analogous to that of Gaussian laws, a direct generalization of the corresponding relations for spherically symmetric distributions and their multivariate analogues. Here two subclasses of invariance hypotheses and criteria for testing them are distinguished by the respective conditions that the asymmetry of the alternative distributions is due either only to biases in the vector m of means or only to asymmetry in the distributions of the random components of X-m. The hypothesis \(H_ 0\) of general form is the conjunction \(H^ 1_ 0\) and \(H^ 2_ 0\) of invariance hypotheses of the first and second indicated types, and a theorem provides a basis for testing \(H_ 0\) by the successive application of criteria for testing the hypotheses \(H^ 2_ 0\) and \(H^ 1_ 0.''\) The proofs of these results hopefully will appear later.
      0 references

      Identifiers