Mixing and moments properties of a non-stationary copula-based Markov process
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS(2020)
摘要
We provide conditions under which a non-stationary copula-based Markov process is geometric beta-mixing and geometric rho-mixing. Our results generalize some results of Beare who considers the stationary case. As a particular case we introduce a stochastic process, that we call convolution-based Markov process, whose construction is obtained by using the C-convolution operator which allows the increments to be dependent. Within this subclass of processes we characterize a modified version of the standard random walk where copulas and marginal distributions involved are in the same elliptical family. We study mixing and moments properties to identify the differences compared to the standard case.
更多查看译文
关键词
Markov process,copula,beta-mixing,rho-mixing,Gaussian process
AI 理解论文
溯源树
样例
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要