A Time-Varying Vine Copula Model for Dependence Analysis of Failure System
作者
Dan Xu,Yuan Jinshuai,Mengli Xing
出处
期刊:2018 International Conference on Sensing,Diagnostics, Prognostics, and Control (SDPC)日期:2018-08-01卷期号:: 437-442被引量:3
标识
DOI:10.1109/sdpc.2018.8664798
摘要
A multivariate approach, called time-varying vine copula, for dependence analysis of complex system is proposed in this paper. The complex system will face the challenges of high-dimensional coupling and changes of dependence overtime. So we proposed an approach that includes the time-varying and vine model, which are used to be described change of parameters with time and to overcome the decomposition of high-dimensional joint distribution respectively. Specifically, three kinds of time-varying parameter model are used in this paper. In short, the aims of this approach are combining vine model and time-varying bivariate copulas, and then extending it to application in failure dependence system.