Association for Public Policy Analysis & Management

In this paper, we propose the combined estimator of full model and post-break estimator, in which combination weight is proportional to the Hausman statistics test, and we derive the optimal weight by minimizing the risk. We derived the exact finite risk for this estimator. Besides, we proof the asymptotic theory and investigate the behavior of our estimator asymptotically. Monte Carlo results proof the accuracy of our estimator in compare to some other existing estimators in the literature.