2019
DOI: 10.2139/ssrn.3508672
|View full text |Cite
|
Sign up to set email alerts
|

Dynamic Effects of Persistent Shocks

Abstract: We provide evidence that many narrative shocks used by prominent literature are persistent. We show that the two leading methods to estimate impulse responses to an independently identified shock (local projections and distributed lag models) treat persistence differently, hence identifying different objects. We propose corrections to re-establish the equivalence between local projections and distributed lag models, providing applied researchers with methods and guidance to estimate their desired object of int… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
8
0

Year Published

2020
2020
2022
2022

Publication Types

Select...
7
1
1

Relationship

2
7

Authors

Journals

citations
Cited by 22 publications
(16 citation statements)
references
References 62 publications
(343 reference statements)
0
8
0
Order By: Relevance
“…It has been argued that, for a true IRF representation, subsequent developments in the shock of interest should be controlled for (Alloza, Gonzalo, and Sanz 2019). Including leading changes in commodity price changes results in larger impacts but does not alter the qualitative channels through which price shocks operate, or the qualitative differences between the transmission channels in each type of commodity exporter.…”
Section: Annex 6c Commodity-driven Productivity Developments: Methodologymentioning
confidence: 99%
“…It has been argued that, for a true IRF representation, subsequent developments in the shock of interest should be controlled for (Alloza, Gonzalo, and Sanz 2019). Including leading changes in commodity price changes results in larger impacts but does not alter the qualitative channels through which price shocks operate, or the qualitative differences between the transmission channels in each type of commodity exporter.…”
Section: Annex 6c Commodity-driven Productivity Developments: Methodologymentioning
confidence: 99%
“…Note that in this special case, the LP estimand is unchanged if we drop all control variables in equation (1). However, the projection coefficient φh differs from the LP (and VAR) estimand if xt correlates with rt or with lags of the data (Alloza, Gonzalo, and Sanz (2019)).…”
Section: Equivalence Between Local Projections and Varsmentioning
confidence: 99%
“…To construct this experiment, I follow Alloza et al. (2019) and include h leads of the interest rate variable in each LP at time h . The authors show that by including leads of a variable, the researcher is able to hold constant any movement in this variable between time t and time t+h$t+h$.…”
Section: Fiscal Policy During Rmentioning
confidence: 99%