The Urbanik generalized convolutions in the non-commutative probability and a forgotten method of constructing generalized convolution (Q692541)

From MaRDI portal
Revision as of 22:40, 5 July 2024 by ReferenceBot (talk | contribs) (‎Changed an Item)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
scientific article
Language Label Description Also known as
English
The Urbanik generalized convolutions in the non-commutative probability and a forgotten method of constructing generalized convolution
scientific article

    Statements

    The Urbanik generalized convolutions in the non-commutative probability and a forgotten method of constructing generalized convolution (English)
    0 references
    0 references
    6 December 2012
    0 references
    Let \(P^+\) denote the set of all probability measures supported on the positive real numbers. A generalized convolution, understood as an associative and commutative binary operation on \(P^+\), satisfying natural algebraic (convex linearity, good behaviour with respect to the dilations, etc.) and topological (separate weak continuity) conditions is a concept introduced and studied in a series of papers in the 1960s--1980s by Urbanik. In particular, in [Bull. Pol. Acad. Sci., Math. 34, 475--486 (1986; Zbl 0609.60013)], \textit{J. Kucharczak} and \textit{K. Urbanik} proposed a notion of weakly stable probability measure (with respect to a given generalized convolution) and showed that such measures can be used to construct another operation on \(P^+\), a so-called weak generalized convolution. In the first part of the reviewed paper, the authors recall the results of Kucharczak and Urbanik concerning the question when a given weak generalized convolution is a genuine generalized convolution and providing an iterative method of constructing weakly stable measures, and, further, the authors apply this method to the case where one starts with the Kendall convolution, obtaining certain explicit formulas for resulting probability kernels. In the second part, the authors propose a modification of the notion of a generalized convolution so that it can be studied for moment sequences (i.e., sequences which arise as moments of a probability measure) as opposed to actual probability measures, prove that the \(q\)-convolution of Carnovale and Koornwinder and the \((q,1)\)-convolution of Kula and Ricard are examples of such generalized convolutions on moment sequences and show that in fact the \(q\)-convolution is a weak generalized convolution with respect to the \((q,1)\)-convolution. Some other examples of relations of the latter type, connecting the deformed convolutions to the classical one, are also listed.
    0 references
    generalized convolution
    0 references
    weakly stable measures
    0 references
    Kendall convolution
    0 references
    moment sequences
    0 references
    q-convolution
    0 references

    Identifiers