Sufficient information in structural VARs
By M. Forni, Luca Gambetti. 2014
232 citations recorded by the source corpus.
Necessary and sufficient conditions under which a VAR contains sufficient information to estimate the structural shocks are derived. On the basis of this theoretical result we propose two simple tests to detect informational deficiency and a procedure to amend a deficient VAR. A simulation based on a DSGE model with fiscal foresight suggests that our method correctly identifies and fixes the informational problem. In an empirical application, we show that a bivariate VAR including unemployment and labor productivity is informationally deficient. Once the relevant information is included into the model, technology shocks appear to be contractionary.
Subfields: Economics and Econometrics
View Paper: https://www.semanticscholar.org/paper/67edde26307d15810c9de64cf2e045dd580c9829
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Economics and Econometrics
- Mikkel Plagborg-Møller
- Luca Gambetti
- M. Forni
- Marco Lippi
- Jushan Bai
- Matteo Barigozzi
- L. Reichlin
- Mark W. Watson
Citations in the corpus
- Bayesian inference on structural impulse response functions (2019, methodology): “Forni et al. (2009) and Forni et al. (2014) use information from large panel data sets to ameliorate the omitted variables problem; based on the same idea, Giannone & Reichlin (2006) and Forni & Gambetti (2014) propose tests of invertibility.”
- Government spending shocks in open economy VARs (2014, methodology): “We test for sufficient information by using the orthogonality test recently proposed by Forni and Gambetti, 2014.”
- Identification and Estimation of Dynamic Causal Effects in Macroeconomics Using External Instruments (2018, unknown): “The focus of this test on the impulse response function – the estimand of interest – differs from existing tests for invertibility, which examine the no-omitted-variables implication by adding variables; see, for example, Forni and Gambetti (2014). Fourth, lest one think that LP-IV is too good to be”