A0524
Title: Identification and estimation of structural vector autoregressive models via LU decomposition
Authors: Masato Shimokawa - Shinshu University (Japan)
Kou Fujimori - Shinshu University (Japan) [presenting]
Abstract: Structural vector autoregressive (SVAR) models are widely used to analyze the simultaneous relationships between multiple time-dependent data. Various statistical inference methods have been studied to overcome the identification problems of SVAR models. However, most of these methods impose strong assumptions for innovation processes, such as the uncorrelation of components. The assumptions for innovation processes are relaxed, and an identification method is proposed for SVAR models under zero restrictions on the coefficient matrices, which correspond to sufficient conditions for LU decomposition of the coefficient matrices of the reduced form of the SVAR models. Moreover, asymptotically normal estimators are established for the coefficient matrices and impulse responses, which enable the construction of test statistics for the simultaneous relationships of time-dependent data. The finite-sample performance of the proposed method is elucidated by numerical simulations. An example of an empirical study that analyzes the impact of policy rates on unemployment and prices is also presented.
