Statistical inference of integer valued time series based on signed thinning operator of symbolic generalized power series in random environment

A kind of integer autoregressive model of random coefficients in random environment based on Signed thinning operator of signed generalized power series thinning operator is studied statistically. Conditional maximum likelihood estimators of model parameters are presented. The simulation results show that the model is effective in solving the parameter estimation problem of negative integer-valued time series in random environment.