Estimation of P(Z < Y) for correlated stochastic time series models

摘要

Let Z and Y represent two time series that are not necessarily independent, and Zn+L, Ym+k denote their values respectively at future times n+L and m+k, where n+L=m+k. Autoregressive (AR), Moving Average (MA), and Autoregressive Moving Average (ARMA) models are employed both under stationary and non-stationary conditions to estimate P(Zn+L