Open AccessEstimation of impulse response functions using long autoregression
Author(s) -
Chang PaoLi,
Sakata Shinichi
Publication year - 2007
Publication title -
the econometrics journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.861
H-Index - 36
eISSN - 1368-423X
pISSN - 1368-4221
DOI - 10.1111/j.1368-423x.2007.00216.x
Subject(s) - estimator , impulse response , asymptotic distribution , autoregressive model , mathematics , vector autoregression , consistency (knowledge bases) , econometrics , parametric statistics , impulse (physics) , statistics , mathematical analysis , geometry , physics , quantum mechanics
Summary This article proposes an alternative methodology to estimate impulse response functions without imposing parametric restrictions. The impulse responses are estimated by regressing the series of interest on estimated innovations, which are the residuals obtained from a prior‐stage ‘long autoregression.’ We establish the consistency and asymptotic normality of the proposed estimator. The proposed estimator is closely related to the estimator of Jordà (2005, American Economic Review 95 , 161–182). Our large sample analysis, as a byproduct, establishes the asymptotic equivalence between Jordà's estimator and our estimator, and provides justifications for the statistical inference method used in Jordà (2005).