A Markov chain on the symmetric group and Jack symmetric functions (Q1191948)

Usually, in the random walks on groups, transitional probability \(\tau(x,y)\) of going from \(y\) to \(x\) depends only on the group element \(xy^{-1}\) and in most cases only on the conjugacy class of \(xy^{-1}\). Diaconis and Shahshahani (see \textit{P. Diaconis} and \textit{M. Shahshahani} [Generating a random permutation with random transpositions, Z. Wahrscheinlichkeitstheor. Verw. Geb. 57, 159-179 (1981; Zbl 0485.60006)]) analyzed a proposed card-shuffling procedure using a random walk on the symmetric group \(S_ f\) where the transitional probability \(t_ 1(\sigma,\pi)\) is \({f\choose 2}^{-1}\) if \(\sigma\pi^{-1}\) is a transposition and 0 otherwise. In their random walk, you can move from a permutation \(\pi\) to any permutation of the form \(\pi(i,j)\) with equal probability. This random walk is denoted by \(W_ f(1)\). In the paper under review the author studies a deformation \(W_ f(\alpha)\) of this Markov chain which is obtained by applying the Metropolis algorithm to \(W_ f(1)\). The stable distribution of \(W_ f(\alpha)\) is \(\alpha^{f- c(\pi)}\) where \(c(\pi)\) denotes the number of cycles of \(\pi\). The main result of the paper is that the eigenvectors of the transition matrix of \(W_ f(\alpha)\) are the Jack symmetric functions. The author uses facts about the Jack symmetric functions due to Macdonald and Stanley to obtain precise estimates for the rate of convergence of \(W_ f(\alpha)\) to its stable distribution.