Dynamic Identification of Dynamic Stochastic General Equilibrium Models
研究了如何从数据的一阶和二阶矩动态识别DSGE模型参数，给出了两类秩条件和阶条件，并考虑了测量误差等情形。
This paper studies dynamic identification of parameters of a dynamic stochastic general equilibrium model from the first and second moments of the data. Classical results for dynamic simultaneous equations do not apply because the state space solution of the model does not constitute a standard reduced form. Full rank of the Jacobian matrix of derivatives of the solution parameters with respect to the parameters of interest is necessary but not sufficient for identification. We use restrictions implied by observational equivalence to obtain two sets of rank and order conditions: one for stochastically singular models and another for nonsingular models. Measurement errors, mean, long-run, and a priori restrictions can be accommodated. An example is considered to illustrate the results.