An Exact Cholesky Decomposition and the Generalized Inverse of the Variance–Covariance Matrix of the Multinomial Distribution, with Applications

Abstract

SUMMARY A symbolic formula is given for the square‐root‐free Cholesky decomposition of the variance–covariance matrix of the multinomial distribution. The evaluation of the symbolic Cholesky factors requires much fewer arithmetic operations than does the general Cholesky algorithm. Since the symbolic formula is not affected by an ill‐conditioned matrix, it is particularly useful when the elements of a probability vector are of quite different orders of magnitude. A simpler formula is obtained for Pederson's procedure of sampling from a multinomial population. An explicit formula of the Moore–Penrose inverse of the variance–covariance matrix is given as well as a symmetric representation of a multinomial density approximation to the multinomial distribution. These formulae facilitate symmetric manipulation of the matrix and are useful in statistical modelling and computation involving the logistic density transformation of the multinomial distribution and in computer simulations of dynamic models in population genetics. Each element of the Cholesky factors is given interesting probabilistic interpretations.

Efficient semiparametric estimation of multi-valued treatment effects under ignorability

“…The form of the matrix H (x, γ ) and Theorem 1 in Tanabe and Sagae (1992) show that H (x, γ ) is symmetric positive definite with 0 < λ min (H (x, γ )) ≤ λ max (H (x, γ )) < 1, which implies that H (x, γ ) ≥ λ min (H (x, γ )) I J and λ min (H (x, γ )) ≥ det (H (x, γ )). These results and the exact Cholesky decomposition of H (x, γ ) give inf x∈X H (x, γ ) ≥ inf x∈X J t=0 L t (g −0 (x, γ )) I J , in a positive semidefinite sense.…”

Asymptotic Least Squares Estimators for Dynamic Games

“…where e is a K × 1-dimensional vector given by e = (1, 1,..., 1) . The expression for the inverse is given for example in Tanabe and Sagae (1992). The inverse of the covariance matrix p is then given by the block-diagonal matrix −1 p = diag −1 p11 , ··· , −1 pN m s .…”

Cooperative Beamforming for Cognitive Radio Systems with Asynchronous Interference to Primary User

“…Since R true is a covariance matrix, it can be factorized using Cholesky decomposition [24]. Therefore, we can write R true = C true C † true , where C true is a lower triangular matrix.…”