The multivariate Laplace-de Moivre theorem
Let \({\mathfrak A}\) denote the vertices of the unit cube in \(R^ k\). A k- dimensional random vector X is said to be Bernoulli if P(X\(\in {\mathfrak A})=1\). A k-dimensional random vector Y is said to have the k-dimensional binomial distribution based on n trials if Y has the same distribution as the sum of n independent identically distributed k-dimensional Bernoulli random vectors. In this paper necessary and sufficient conditions for the convergence of (matrix) normed and centered k-variate binomial vectors to a full k- variate normal law are found. The norming matrices are also characterized. A converse to the univariate Laplace-De Moivre theorem is also obtained.
This page was built for publication: The multivariate Laplace-de Moivre theorem
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1070644)