Non-causality In Bivariate Binary Time Series

In this paper, we discuss Granger non-causality in bivariate binary time series ,con-struct a dynamic discrete-time bivariate probit modle , in which causality between the bivariate variable can be discussed. Granger non-causality with the bivariate probit modle is discussed . The modle with exogenous processes ,and the generalized Markov dynamic modle is descussed . The conditions for Granger non-causality under those assumption is work out. The proposed model can be estimated by Maximum Likelihood. The properties of the modle for homogeneous population is discussed. The parameter estimates are worked out by the Newton-Raphson algorithm . Example is analyzed by the proposed method.