【主讲】Denis Tkachenko 助教授 (新加坡国立大学)
【主题】Local and Global Parameter Identification in DSGE Models Allowing for Indeterminacy
【时间】2013年9月25日 (周三) 15：30-17：00
【地点】上海财经大学经济学院楼710室
【语言】英文
【摘要】This paper presents a unified framework for analyzing local and global identification in log linearized DSGE models that encompasses both determinacy and indeterminacy. The analysis is conducted from a frequency domain perspective. First, for local identification, it presents necessary and sufficient conditions for the identification of the structural parameters along with the sunspot parameters, the identification of the former irrespective of the latter and the iden- tification of the former conditional on the latter. The conditions apply to both singular and nonsingular systems and permit analysis using a subset of frequencies. Second, for global iden- tification, the paper utilizes the Kullback-Leibler distance between two stationary vector linear processes and shows that global identification fails if and only if the minimized distance equals zero. Consequently, it delivers a set of parameter values that yield observational equivalence under identification failure. The global identification condition requires a nonsingular system but can be applied across models with different structures. Third, to understand the strength of identification, the paper proposes a measure of the empirical distance between two DSGE models. It gauges the possibility of distinguishing one model from another using likelihood ratio tests based on a infinite number of observations generated by the models. Finally, although the theory is developed under a DSGE setup, it is applicable to other dynamic linear models with well defined spectra such as structural (factor augmented) vector autoregressive moving average Models.
