Indirect inference under stochastic restrictions

Abstract

This paper is focused on the stochastic restriction approach, the NLS and the indirect inference estimators. A methodology to combine sample and prior information is suggested when the indirect inference estimation method is in use. This goal is achieved through stochastic restrictions approach. The resulting estimator is proved to be more efficient than the indirect inference estimator under specific assumptions about the behaviour of the stochastic restriction. As an illustration of the proposed methodology, the capital stock of an economy is estimated through the estimation of its stochastic rate of the depreciation.