Approximation orders in the conditional central limit theorem for weakly dependent random variables (Q1281604): Difference between revisions

The author studies the speed of convergence in conditional central limit theorems for weakly dependent sequences. Specifically, assume that \((X_k)_{k\geq 1}\) is strictly stationary and strongly mixing with partial sums \(S_n: =\sum^n_{k=1} X_k\). Then the convergence rate of \(\sup_{t\in\mathbb{R}}|P(S_n/ \sqrt{\text{Var}(S_n)}\leq t\mid B)- \Phi(t) |\) is estimated from above, under conditions on the set \(B\), on the mixing coefficients and on the moments of the process \((X_k)_{k\geq 1}\). The present paper generalizes results of \textit{D. Landers} and \textit{L. Rogge} [Z. Wahrscheinlichkeitstheorie Verw. Geb. 66, 227-244 (1984; Zbl 0525.60029)] for independent random variables and of the author [Vietnam J. Math. 21, 13-19 (1993)] for uniformly mixing processes.