The Large Deviation Theory of Importance Sampling Estimators
In this section we first prove a very general theorem regarding the variance rate of importance sampling estimators. At first sight the setting may appear to be fairly abstract but we argue that the level of generality presented here will pay dividends later on. We encourage the reader to follow the problem setup given here with some care and then compare it with the first example given below to see how this framework is actually used.
KeywordsMarkov Chain Importance Sampling Moment Generate Function Simulation Distribution Asymptotic Rate
Unable to display preview. Download preview PDF.