Accelerated Variance Reduction Methods on GPU
Department of Quantitative Finance, National Tsing-Hua University, Hsinchu, Taiwan, R.O.C.
National Tsing-Hua University, 2014
@article{han2014accelerated,
title={Accelerated Variance Reduction Methods on GPU},
author={Han, Chuan-Hsiang and Lin, Yu-Tuan},
year={2014}
}
Monte Carlo simulations have become widely used in computational finance. Standard error (SE in short) is the basic notion to measure the quality of a Monte Carlo estimator, and the square of SE is defined as the variance divided by the total number of simulations. Variance reduction methods have been developed as efficient algorithms by means of probabilistic analysis. GPU acceleration plays the role of increasing the total number of simulations. We show that the total effect of combining variance reduction methods as software algorithms with GPU acceleration as hardware device can yields a tremendous speed up for financial applications such as option pricing and probability estimation of joint default.
September 15, 2014 by hgpu
