Approximation of quadratic forms of independent random vectors by accompanying laws
Let \(Q\) be a linear symmetric and bounded operator mapping \(\mathbb{R}^d\) into \(\mathbb{R}^d\). Let us put \(Q[x]= \langle Qx,x\rangle\), \(x\in\mathbb{R}^d\). Assume \(Q\) is non-degenerate. Let \(\{X, X_1, X_2,\dots\}\) and \(\{Y, Y_1, Y_2,\dots\}\) be sequences of random vectors taking values in \(\mathbb{R}^d\) and such that \(L(Y)= e(L(X))\). Assume \(X\) is not concentrated on a proper closed linear subspace of \(\mathbb{R}^d\). Let us put \(S_N= N^{-1/2}(X_1+\cdots+ X_n)\), \(T_N= N^{-1/2}(Y_1+\cdots+ Y_N)\). The authors investigate the rate of approximation of the distribution of the sum of i.i.d. random vectors by the accompanying laws on the balls, ellipsoids and hyperboloids. The main result gives the following theorem: Assume that \(9\leq d\leq\infty\). Let \(EX= 0\), and \(E| X|^{8/3}<\infty\). Then \[ \sup_x| P\{Q[S_N- a]< x\}- P\{Q[T_N- a]< x\}|= O((1+| a|^4) N^{-1}), \] and the constant in this bound depends on \(E| X|^{8/3}\), on \(Q\) and the covariance operator of \(X\) only. The authors also prove that the presented rate of convergence cannot be faster.
- scientific article; zbMATH DE number 29784
- scientific article; zbMATH DE number 4064247
- Two properties of vectors of quadratic forms in Gaussian random variables
- Approximations to Joint Distributions of Definite Quadratic Forms
- L_p-approximable sequences of vectors and limit distribution of quadratic forms of random variables
- Multivariate quadratic forms of random vectors
- scientific article; zbMATH DE number 54139
- Uniform rates of convergence in the CLT for quadratic forms in multidimensional spaces
- Optimal bounds in non-Gaussian limit theorems for U-statistics
- Compound Poisson approximation
- Explicit rates of approximation in the CLT for quadratic forms
- Approximation of a sample by a Poisson point process
- Compound Poisson approximations for symmetric vectors
- Multidimensional version of the results of Komlos, Major and Tusnady for vectors with finite exponential moments
- Random vectors satisfying Khinchine–Kahane type inequalities for linear and quadratic forms
This page was built for publication: Approximation of quadratic forms of independent random vectors by accompanying laws
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3842397)
