Variance Reduction in Monte Carlo Estimators via Empirical Variance Minimization
- Authors: Belomestny D.V.1,2, Iosipoi L.S.1,3, Zhivotovskiy N.K.1,2,4
-
Affiliations:
- National Research University Higher School of Economics
- University of Duisburg-Essen
- Faculty of Mechanics and Mathematics
- Institute for Information Transmission Problems
- Issue: Vol 98, No 2 (2018)
- Pages: 494-497
- Section: Mathematics
- URL: https://journals.rcsi.science/1064-5624/article/view/225565
- DOI: https://doi.org/10.1134/S1064562418060261
- ID: 225565
Cite item
Abstract
For Monte Carlo estimators, a variance reduction method based on empirical variance minimization in the class of functions with zero mean (control functions) is described. An upper bound for the efficiency of the method is obtained in terms of the properties of the functional class.
About the authors
D. V. Belomestny
National Research University Higher School of Economics; University of Duisburg-Essen
Email: iosipoileonid@gmail.com
Russian Federation, Moscow; Duisburg and Essen
L. S. Iosipoi
National Research University Higher School of Economics; Faculty of Mechanics and Mathematics
Author for correspondence.
Email: iosipoileonid@gmail.com
Russian Federation, Moscow; Moscow
N. K. Zhivotovskiy
National Research University Higher School of Economics; University of Duisburg-Essen; Institute for Information Transmission Problems
Email: iosipoileonid@gmail.com
Russian Federation, Moscow; Duisburg and Essen; Moscow