Bayesian inference on structural impulse response functions

Plagborg‐Møller, Mikkel

doi:10.3982/qe926

Cited by 35 publications

(21 citation statements)

References 75 publications

(101 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One approach is to assume that the shocks are independent and non‐Gaussian, and to exploit higher order moment restrictions to identify the causal structure (Lanne and Saikkonen, ; Gospodinov and Ng, ; Gouriéroux et al ., ). A second approach is to use a priori informative priors (Plagborg‐Møller, ). Here, we have shown that there is a third approach, which is to use an external instrument.…”

Section: Discussionmentioning

confidence: 99%

“…Beaudry et al . () and Plagborg‐Møller () provide numerical evidence that in many cases the non‐invertible (non‐fundamental) representation of a time series may be very close to its invertible representation. With this motivation, we focus on non‐invertible IRFs that represent small departures from an invertible null.…”

Section: A Test Of Invertibilitymentioning

confidence: 99%

“…In addition to establishing the null distribution of the test, we wish to examine its distribution under an alternative to check that the test has power against noninvertibility. Beaudry et al (2015) and Plagborg-Møller (2016b) provide numerical evidence that in many cases the non-invertible (non-fundamental) representation of a time series may be very close to its invertible representation. With this motivation, we focus on non-invertible IRFs that represent small departures from an invertible null.…”

Section: Null and Local Alternativementioning

confidence: 99%

See 2 more Smart Citations

Identification and Estimation of Dynamic Causal Effects in Macroeconomics Using External Instruments

Stock

Watson

2018

The Economic Journal

363

178

View full text Add to dashboard Cite

External sources of as‐if randomness — that is, external instruments — can be used to identify the dynamic causal effects of macroeconomic shocks. One method is a one‐step instrumental variables regression (local projections – IV); a more efficient two‐step method involves a vector autoregression. We show that, under a restrictive instrument validity condition, the one‐step method is valid even if the vector autoregression is not invertible, so comparing the two estimates provides a test of invertibility. If, however, lagged endogenous variables are needed as control variables in the one‐step method, then the conditions for validity of the two methods are the same.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: A Test Of Invertibilitymentioning

confidence: 99%

Section: Null and Local Alternativementioning

confidence: 99%

See 1 more Smart Citation

Identification and Estimation of Dynamic Causal Effects in Macroeconomics Using External Instruments

Stock

Watson

2018

The Economic Journal

363

178

View full text Add to dashboard Cite

show abstract

“…Similar to what is done in the (Structural) VAR literature to compute confidence intervals for impulse response functions, one can quantify the uncertainty around the Wold components by, for example, bootstrapping the data using the estimated VAR parameters and the fitted residuals. Alternatively, one could conduct inference directly on the moving average representation of the data (without estimation and inversion of the VAR) along the line of Barnichon and Matthes () and Plagborg‐Moller (). We view this as an interesting avenue for future research.…”

Section: Discussionmentioning

confidence: 99%

A persistence‐based Wold‐type decomposition for stationary time series

Ortu¹,

Severino

Tamoni³

et al. 2020

View full text Add to dashboard Cite

This paper shows how to decompose weakly stationary time series into the sum, across time scales, of uncorrelated components associated with different degrees of persistence. In particular, we provide an Extended Wold Decomposition based on an isometric scaling operator that makes averages of process innovations. Thanks to the uncorrelatedness of components, our representation of a time series naturally induces a persistence-based variance decomposition of any weakly stationary process. We provide two applications to show how the tools developed in this paper can shed new light on the determinants of the variability of economic and financial time series.

show abstract

“…Macroeconomists now consider various types of news shocks, beginning with TFP news (Beaudry and Portier (2006)), and including government spending news (Ramey (2011)) and monetary policy news (Campbell, Evans, Fisher, and Justiniano (2012)). Plagborg‐Møller (2019) estimated point‐identified IRFs with news shocks directly using a prior on shapes. We use ranking restrictions to condition on shocks that lead to a TFP build‐up for TFP news shocks (the first shock) and mean‐reversion for TFP surprises:

\begin{matrix} trueright left \\ - λ \times r_{a_{1}, TFP}^{0} & > - r_{a_{1}, TFP}^{1} > - λ \times r_{a_{1}, TFP}^{2} with λ = 1, \\ λ \times r_{a_{2}, TFP}^{0} & > r_{a_{2}, TFP}^{1} > λ \times r_{a_{2}, TFP}^{2} with λ = 1 . \end{matrix}

These restrictions rule out aliasing of news and surprise shocks.…”

Section: Modelmentioning

confidence: 99%

Identification and inference with ranking restrictions

Amir-Ahmadi

Drautzburg

2021

View full text Add to dashboard Cite

We propose to add ranking restrictions on impulse‐responses to sign restrictions to narrow the identified set in vector autoregressions (VARs). Ranking restrictions come from micro data on heterogeneous industries in VARs, bounds on elasticities, or restrictions on dynamics. Using both a fully Bayesian conditional uniform prior and prior‐robust inference, we show that these restrictions help to identify productivity news shocks in the data. In the prior‐robust paradigm, ranking restrictions, but not sign restrictions alone, imply that news shocks raise output temporarily, but significantly. This holds both in an application with rankings in the form of heterogeneity restrictions and in another applications with slope restrictions as rankings. Ranking restrictions also narrow bounds on variance decompositions. For example, the bound of the contribution of news shocks to the forecast error variance of output narrows by about 30 pp at the one‐year horizon. While misspecification can be a concern with added restrictions, they are consistent with the data in our applications.

show abstract

Bayesian inference on structural impulse response functions

Cited by 35 publications

References 75 publications

Identification and Estimation of Dynamic Causal Effects in Macroeconomics Using External Instruments

Identification and Estimation of Dynamic Causal Effects in Macroeconomics Using External Instruments

A persistence‐based Wold‐type decomposition for stationary time series

Identification and inference with ranking restrictions

Contact Info

Product

Resources

About