$L^p$ bounds for a central limit theorem with involutions

Abstract

Let $E=((e_{ij}))_{n\times n}$ be a fixed array of real numbers such that $e_{ij}=e_{ji}, e_{ii}=0$ for $1\le i,j \le n$. Let the permutation group be denoted by $S_n$ and the collection of involutions with no fixed points by $\Pi_n$, that is, $\Pi_n=\{\pi\in S_n: \pi^2= id, \pi(i)\neq i\,\forall i\}$ with id denoting the identity permutation. For $\pi$ uniformly chosen from $\Pi_n$, let $Y_E=\sum_{i=1}^n e_{i\pi(i)}$ and $W=(Y_E-\mu_E)/\sigma_E$ where $\mu_E=E(Y_E)$ and $\sigma_E^2= Var(Y_E)$. Denoting by $F_W$ and $\Phi$ the distribution functions of $W$ and a $\mathcal{N}(0,1)$ variate respectively, we bound $||F_W-\Phi||_p$ for $ 1\le p\le \infty$ using Stein's method and the zero bias transformation. Optimal Berry-Esseen or $L^\infty$ bounds for the classical problem where $\pi$ is chosen uniformly from $S_n$ were obtained by Bolthausen using Stein's method. Although in our case $\pi \in \Pi_n$ uniformly, the $L^p$ bounds we obtain are of similar form as Bolthausen's bound which holds for $p=\infty$. The difficulty in extending Bolthausen's method from $S_n$ to $\Pi_n$ arising due to the involution restriction is tackled by the use of zero bias transformations.