ABSTRACT
Consider using simulation to estimate the parameters of a distribution. Due to the inherent randomness, our estimates will vary from one simulation run to the other. Naturally it is desirable to reduce this variability as much as possible, thus improving the reliability of the resulting estimates. We can always reduce overall variability by increasing the number of simulation trials, but this can take too long. The variance reduction techniques we introduce sometimes offer substantial efficiency gains by working smarter not harder, using an understanding of the structure of the simulation task at hand.
