On making use of a result of Imhof, an integral representation of the distribution function of linear combinations of the components of a Dirichlet random vector is obtained. In fact, the distributions of several statistics such as Moran and Geary's indices, the Cliff‐Ord statistic for spatial correlation, the sample coefficient of determination, F‐ratios and the sample autocorrelation coefficient can be similarly determined. Linear combinations of the components of Dirichlet random vectors also turn out to be a key component in a decomposition of quadratic forms in spherically symmetric random vectors. An application involving the sample spectrum associated with series generated by ARMA processes is discussed.

No abstract

This paper provides computable representations for the evaluation of the probability content of cones in isotropic random fields. A decomposition of quadratic forms in spherically symmetric random vectors is obtained and a representation of their moments is derived in terms of finite sums. These results are combined to obtain the distribution function of quadratic forms in spherically symmetric or central elliptically contoured random vectors. Some numerical examples involving the sample serial covariance are provided. Ratios of quadratic forms are also discussed. 1998Academic Press

No abstract

Young-Ho CheongBank One, Wilmington, U.S.A. SYNOPTIC ABSTRACTA recursive relationship for determining the moments of a quadratic form in normal variables as well as an explicit formula for approximating a continuous density function defined on a compact support from its moments are derived in this paper. Each of these results have, on their own, a plethora of applications as quadratic forms are ubiquitous in Statistics and the moments of most test statistics that are confined to closed intervals can be readily evaluated; they are combined herewith to produce an approximation to the null distribution of the Durbin-Watson statistic, which for all intents and purposes, can be viewed as exact. The proposed approach takes into account the observation matrix of explanatory variables associated with the assumed regression model, and more accuracy can always be gained by making use of additional moments.Furthermore, the Durbin-Watson statistic is shown to be invariant in the class of spherically distributed error vectors, and an integral formu la is derived for evaluating its moments under the assumption that the error vector has a general covariance structure. A numerical example illustrates the proposed methodology.