Monomial-factor conjecture for the stationary distribution

About 15 years old · traced to

Let SnS_n be the symmetric group and let ζ(w)\zeta(w) be the normalized stationary-distribution component defined by the nullspace of PT−IP^T-I. Assume that each ζ(w)\zeta(w) is a polynomial, and let η(w)\eta(w) denote the largest monomial that can be factored out of ζ(w)\zeta(w). Let [m]={0,1,2,…,m}[m]=\{0,1,2,\ldots,m\}. Monomial-factor conjecture. The map w↦η(w)w\mapsto\eta(w) is an nn-to-11 map from SnS_n to

{x1a1+a2+⋯+an−2x2a2+⋯+an−2⋯xn−2an−2∣(a1,a2,…,an−2)∈[n−2]×[n−3]×⋯×[1]}.\left\{x_1^{a_1+a_2+\cdots+a_{n-2}}x_2^{a_2+\cdots+a_{n-2}}\cdots x_{n-2}^{a_{n-2}}\mathrel{|}(a_1,a_2,\ldots,a_{n-2})\in[n-2]\times[n-3]\times\cdots\times[1]\right\}.

Moreover, if

η(w)=x1a1+a2+⋯+an−2x2a2+⋯+an−2⋯xn−2an−2,\eta(w)=x_1^{a_1+a_2+\cdots+a_{n-2}}x_2^{a_2+\cdots+a_{n-2}}\cdots x_{n-2}^{a_{n-2}},

then

ai=#{k∈[i+2,n]∣wk∈[wi,wi+1]},a_i=\#\{k\in[i+2,n]\mid w_k\in[w_i,w_{i+1}]\},

where [wi,wi+1][w_i,w_{i+1}] denotes a cyclic subinterval of [n][n].

This gives a proposed explicit classification of the monomial factors of the stationary-distribution components; the source does not state whether it has been proved or disproved.

References

Primary source

Thomas Lam and Lauren Williams, “A Markov chain on the symmetric group which is Schubert positive?”, arXiv:1102.4406 (2011).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.