Analysis on adjoint non-recurrent property of nonlinear time series in random environment domain

By introducing a random interference into the typical of nonlinear time series model, this paper establishes a RENLAR model: $$X_{n+1}=T(X_n)+ e_{n+1}(Z_{n+1})$$ . The author introduces the definition of adjoint non-recurrence, and utilizing general state space Markov chain theorem, we obtain some criteria for non-recurrence and adjoint non-recurrence of nonlinear time series models in random environment domain and analyze adjoint non-recurrence of some models by using these criteria. Copyright Springer-Verlag 2007