scientific article; zbMATH DE number 4119441

From MaRDI portal
Publication:4733258





[For the entire collection see Zbl 0672.00008.] Consider a process \(X_ 1,...,X_ n\). There are many important statistics based on \(X_ 1,...,X_ n\), for which it is very difficult or impossible to derive their distributions, even if the process is completely specified. For example, if \(\xi_ n=\max (X_ 1,...,X_ n),\) then only some limit and asymptotic properties of its distribution are known and in many cases it is necessary to assume that the process is normal. If explicit formulas for such a distribution are not known, it remains the possibility to simulate the process and to find empirical distributions of the considered statistics. The general problem is how to simulate a stationary process with a given covariance function and with a given one-dimensional marginal distribution. \textit{M. M. Sondhi} [Bell. Syst. Tech. J. 62, 679-701 (1983)] proposed a method based on an expansion into a power series using Hermite polynomials. In our paper we demonstrate another procedure which can be called a moment method.











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4733258)