How to improve a random number generator

It is demonstrated that the one-dimensional random walk is sensitive both t o short and long correlation of random numbers. In the case of correlated r andom numbers, the walker is trapped in the vicinity of the correlation. Th e result is a multipeak range distribution of the walker. Hence, the one-di mensional random walk can be used as a supplement to the standard tests for revealing long correlations of the generated random numbers. By implementi ng variety of tests to several generators we have shown that the quality ca n be improved by using a double randomization technique only if the random numbers, to be sampled from, an of a good quality. The sampling generator i s less important. We propose, as well, a generator that produces high-quality random numbers for many stochastic processes. (C) 2000 Elsevier Science B.V. All rights re served.