A COMPARISON OF VECTORIZABLE DISCRETE SAMPLING METHODS IN MONTE-CARLOAPPLICATIONS

The performance of various vectorizable discrete random-sampling metho ds, along with the commonly used inverse sampling method, is assessed on a vector machine. Monte Carlo applications involving, one-dimension al, two-dimensional and multi-dimensional probability tables are used in the investigation. Various forms of the weighted sampling method an d methods that transform the original probability table are examined. It is found that some form of weighted sampling is efficient, when the original probability distribution is not far from uniform or can be a pproximated analytically. Table transformation methods, though requiri ng additional memory storage, are best suited in applications where mu ltidimensional tables are involved.