Multinomial and Krawtchouk Approximations to the Generalized Multinomial Distribution

From MaRDI portal





Let \(S_n\) be a sum of independent but not identically distributed Bernoulli random vectors \(X_1,\dots,X_n\) in \(R^k\). In this paper, the approximation of \({\mathcal L}(S_n)\) by a multinomial distribution is considered, and error bounds provided in both total variation and \(l_\infty\) norms. The bounds improve on those of \textit{W.-L. Loh} [Ann. Appl. Probab. 2, No. 3, 536-554 (1992; Zbl 0759.62007)]. Finer approximations using appropriate signed measures are also derived. The approach is an interesting multivariate extension of the author's paper, Theory Probab. Appl. 45, No. 2, 258-272 (2000; Zbl 0984.62008)].











This page was built for publication: Multinomial and Krawtchouk Approximations to the Generalized Multinomial Distribution

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4328526)