Title: Pair-copula estimation of distribution algorithms
Authors: Huimin Gao; Xiaoping Wang
Addresses: Mechanical and Electrical Engineering College, Jiaxing University, Jiaxing, Zhejing, 314001, China ' School of Business, Jiaxing University, Jiaxing, Zhejing, 314001, China
Abstract: Copula theory provides a promising solution for the estimation of population probability in estimation distribution algorithms (EDAs), and more and more researchers pay attention to copula-EDAs. Most of the copula-EDAs researches are related to two variables case, in this paper, by taking advantage of the ability of pair-copula in high-dimensional correlation construction, a new algorithm is proposed, called pair-copula estimation distribution algorithms (pcEDAs). The architecture of pcEDAs is provided, and sampling method of the probability model is discussed, the simulation results based on two different vines, namely C-vine and D-vine, show that the proposed algorithm is not only feasible, but also perform very well.
Keywords: pair-copula construction; PCC; vines; estimation of distribution algorithms; EDA; copula theory; population probability; modelling; simulation.
International Journal of Computing Science and Mathematics, 2013 Vol.4 No.2, pp.186 - 196
Received: 17 Apr 2013
Accepted: 26 Apr 2013
Published online: 10 May 2014 *