Bourbaki

Exercise 42, § 1

Tag 02LI, permanent. Cite it as [Bourbaki, Tag 02LI].

Spectral Theories, II, p. 282

Let n⩾2n\geqslant 2 be an integer. We denote by G the group SnS_n endowed with its Haar measure μn\mu_n of total mass 1. Let ϖ:Sn→R\varpi : S_n\rightarrow \mathbf{R} be the mapping which associates to σ\sigma the number a+ba+b, where aa is the number of cycles in the decomposition of σ\sigma into a product of disjoint cycles (A, I, p. 60, prop. 7) and bb is the number of fixed points of σ\sigma. We denote by νn\nu_n the image measure νn=ϖ(μn)\nu_n=\varpi (\mu_n) on R\mathbf{R}. It is a probability measure.

a) The characteristic function φn\varphi_n of νn\nu_n satisfies

φn(t)=∏n(1−1j+eitj) \varphi_n(t) =\prod^n(1-\frac{1}{j}+\frac{e^{it}}{j})
j=1j=1

for every t∈Rt\in \mathbf{R}.

b) Let ϖn\varpi_n be a Poisson law of parameter log(nn) and ψn\psi_n its characteristic function. When n→+∞n\rightarrow +\infty, one has

φn(t)=1Γ(it)ψn(t)(1+o(1)) \varphi_n(t) =\frac{1}{\Gamma(it)}\psi_n(t)(1 +o(1))

uniformly for t∈Rt\in \mathbf{R}. (Use Gauss's formula for the gamma function, cf. FVR, VII, p. 11, formula (3).)

c) Let fn(x)=(x−f_n(x) = (x- log(nn))/√/\surdlog(nn) for x∈Rx\in \mathbf{R}. The sequence of measures fn(μn)f_n(\mu_n) converges tightly to the standard Gaussian law when n→+∞n\rightarrow +\infty.

Solution

No solution has been written yet.

Cited by

Nothing.

Cites

Nothing.

Machine translation of the English, by gpt-5-6. Not checked by a person.