Perfect simulation of autoregressive models with infinite memory
Probability
2015-06-11 v1 Mathematical Physics
math.MP
Abstract
In this paper we consider the problem of determining the law of binary stochastic processes from transition kernels depending on the whole past. These kernels are linear in the past values of the process. They are allowed to assume values close to both 0 and 1, preventing the application of usual results on uniqueness. More precisely we give sufficient conditions for uniqueness and non-uniqueness. In the former case a perfect simulation algorithm is also given.
Cite
@article{arxiv.1210.7347,
title = {Perfect simulation of autoregressive models with infinite memory},
author = {Emilio De Santis and Mauro Piccioni},
journal= {arXiv preprint arXiv:1210.7347},
year = {2015}
}
Comments
12 pages