Importance sampling of heavy-tailed iterated random functions
Abstract
We consider a stochastic recurrence equation of the form , where , and is an i.i.d. sequence of positive random vectors. The stationary distribution of this Markov chain can be represented as the distribution of the random variable . Such random variables can be found in the analysis of probabilistic algorithms or financial mathematics, where would be called a stochastic perpetuity. If one interprets as the interest rate at time , then is the present value of a bond that generates unit of money at each time point . We are interested in estimating the probability of the rare event , when is large; we provide a consistent simulation estimator using state-dependent importance sampling for the case, where is heavy-tailed and the so-called Cram\'{e}r condition is not satisfied. Our algorithm leads to an estimator for . We show that under natural conditions, our estimator is strongly efficient. Furthermore, we extend our method to the case, where is defined via the recursive formula and is a sequence of i.i.d. random Lipschitz functions.
Cite
@article{arxiv.1609.03182,
title = {Importance sampling of heavy-tailed iterated random functions},
author = {Bohan Chen and Chang-Han Rhee and Bert Zwart},
journal= {arXiv preprint arXiv:1609.03182},
year = {2016}
}