RT Journal Article
T1 Testing for fundamental vector moving average representations
A1 Chen, Bin
A1 Choi, Jinho
A1 Escanciano, Juan Carlos
AB We propose a test for invertibility or fundamentalness of structural vector autoregressivemoving average models generated by non-Gaussian independent andidentically distributed structural shocks. We prove that in these models and undersome regularity conditions the Wold innovations are a martingale differencesequence (mds) if and only if the structural shocks are fundamental. This simplebut powerful characterization suggests an empirical strategy to assess invertibility.We propose a test based on a generalized spectral density to check for themds property of the Wold innovations. This approach does not require the specificationand estimation of the economic agent’s information flows or the identificationand estimation of the structural parameters and the noninvertible roots.Moreover, the proposed test statistic uses all lags in the sample and it has a convenientasymptotic N(0  1) distribution under the null hypothesis of invertibility,and hence, it is straightforward to implement. In case of rejection, the test can befurther used to check if a given set of additional variables provides sufficient informationalcontent to restore invertibility. A Monte Carlo study is conducted toexamine the finite-sample performance of our test. Finally, the proposed test isapplied to two widely cited works on the effects of fiscal shocks by Blanchard andPerotti (2002) and Ramey (2011).
PB Wiley
SN 1759-7323
YR 2017
FD 2017-03-29
LK https://hdl.handle.net/10016/35096
UL https://hdl.handle.net/10016/35096
LA eng
NO The research of Escanciano was funded by the Spanish Plan Nacional de I+D+I, referencenumber ECO2014-55858-P.
DS e-Archivo
RD 27 jul. 2024