We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Variations Detection of Bivariate Dependence Based on Copulas Model.
- Authors
Yan-ling Li; Yun-peng Zhang; Ling Zhou
- Abstract
Copulas theory provides a convenient way to express joint distributions of two random variables. This paper presents an introduction to detect variations for bivariate relationship, based on the methods of sliding window and copula function, illustrated by the case of monthly precipitation and streamflow sequence in Xianyang station of Weihe River nearly 60. Euclidean distance criterion is presented for selecting an appropriate model, estimating its parameters, and checking its goodness-of-fit. We obtain the combined probability density copula function of relationship between precipitation and streamflow before and after the variations. Result shows that the relationship variant between precipitation and streamflow occurred in 1993, and at different times, precipitation and streamflow follow the different probability distribution function. The goal of this paper is to put forward copula-based in the field of variations detection, so as to provide a stepping stone exploring variations detection research of the bivariate dependence.
- Subjects
PROBABILITY density function; BIVARIATE analysis; DEPENDENCE (Statistics); PARAMETER estimation; EUCLIDEAN distance
- Publication
IAENG International Journal of Applied Mathematics, 2017, Vol 47, Issue 3, p255
- ISSN
1992-9978
- Publication type
Article