Empirical likelihood-based inference in generalized random coefficient autoregressive model with conditional moment restrictions

This paper concentrates on the parameter estimating strategy for the generalized random coefficient autoregressive (GRCA) model in the presence of the auxiliary information. We propose a weighted least squares estimate for the model parameters and empirical likelihood (EL) based weights are obtained through using these auxiliary information. The asymptotic distribution of our proposed estimator is normal distribution and the asymptotic variance is reduced compared to the least square (LS) estimator. Therefore, our method yields more efficient estimates. We also carry out some simulation experiments to assess the performance of the suggested estimator and illustrate the usefulness of this method through the analysis of a real time series data sets.