226
Appendix B
II. GENERATING RANDOM VARIATES
Most computer programming languages are capable of generating a random
variate from a uniform distribution where every real number on an interval
0 1).
The random number generator uses a seed to initialize the process. If
the user does not provide a value for the seed, the computer will use it’s
internal clock for initialization. Each call to the random number generator
then creates a random number and a new seed for the next call.
Computer generated random numbers are commonly referred to as
pseudo-random variates since the computer will generate the same sequence
of numbers, if it is given the same seed for initialization.
Unless the user knows how the computer arrives at its internal seed, it
is strongly suggested that an external seed be used. The last generated seed is
then saved and used to initialize the random number generator on the next
application.
III. GENERATING RANDOM VARIATES FROM OTHER
DISTRIBUTIONS
Many algorithms for generating other random variates from a uniform random variate are widely available in the literature. A collection of usable
algorithms can be found in Rubenstein [B1].
REFERENCES
B1. Rubenstein RY. Simulation and the Monte Carlo Method, John Wiley and
Sons, 1981.