Variance Reduction Techniques
In Chap. 1, we said that one way of improving the Monte Carlo integration error is to try reducing the variance σ2 of the integrand f. More precisely, the goal is to find another function Ø whose integral is equal to the integral of f but whose variance is smaller than that of f. Methods that achieve this are called variance reduction techniques, and we will be describing several of them in this chapter. This topic has been widely studied and is surveyed, for example, in [45, 165, 243, 247, 321, 391], which also give several other references.
KeywordsService Time Importance Sampling Variance Reduction Monte Carlo Estimator Interarrival Time
Unable to display preview. Download preview PDF.