Abstract
A truncation-rejection method for the generation of random variates from the Erlang distribution is presented. A procedure using the Kolmogorov-Smirnov statistic is developed for the comparison of exact and approximate methods for random variate generation. This procedure is used to compare the efficiency of the truncation rejection method against the well known summation method for Erlang variates. The rejection method is found to be mere efficient than the summation method for large values of the Erlang shape parameter
†This work was supported by National Science Foundation Grant GJ–33764X
†This work was supported by National Science Foundation Grant GJ–33764X
Notes
†This work was supported by National Science Foundation Grant GJ–33764X