SIMULATION OF DISCRETE UNIVARIATE AND MULTIVARIATE PROBABILITY DISTRIBUTIONS

Abstract

Simulation describes a wealth of varied and useful techniques, all connected with the mimicking of the rules of a model of some kind. (Shannon, 1975). In this article an exploratory technique for the simulation of discrete univariate and multivariate distribution is proposed. The technique used only the random number table (Scott & Lindley, 1984). The uniform random numbers are used for simulation of randon experiment which would be modelled by the discrete probability distribution with certain parameters. In this article it is shown that the technique may be used to simlate the following distributions: namely, Bernoulli, Binomial, Negative binomial, Geometric & Hypergeometric (Hogg and Craig, 1978). However, in the examples, only binomial, geometric and trinomial are discussed. The adequacy of the model was tested for the empirical probability distribution by using the Chi-square test statistics (Hogg & Craig, 269-1978). The exploratory method is also used to judge the quality of simulated values of random variables graphically (Hoaglin, Mosteller&Tukey, 1985).