Two algorithms are presented to generate random number samples from a Poisson distribution. The first algorithm exploits the relationship between the exponential distribution and the Poisson distribution. It is simple to implement, but its performance varies as a function of the mean of the distribution. The second algorithm uses a combination of techniques to efficiently generate samples in a time that is independent of the mean. This algorithm is found to be about ten times faster than the first algorithm for a particular set of parameters.
|Original language||English (US)|
|Number of pages||11|
|Journal||Transactions of the Society for Computer Simulation|
|State||Published - Jan 1 1989|