Local projection inference in high dimensions

Robert Adamek; Stephan Smeekes; Ines Wilms

doi:10.1093/ectj/utae012

Local projection inference in high dimensions

Robert Adamek^*, Stephan Smeekes, Ines Wilms

^*Corresponding author for this work

Research output: Contribution to journal › Article › Academic › peer-review

Abstract

In this paper, we estimate impulse responses by local projections in high-dimensional settings. We use the desparsified (de-biased) lasso to estimate the high-dimensional local projections, while leaving the impulse response parameter of interest unpenalized. We establish the uniform asymptotic normality of the proposed estimator under general conditions. Finally, we demonstrate small sample performance through a simulation study and consider two canonical applications in macroeconomic research on monetary policy and government spending.

Original language	English
Pages (from-to)	323-342
Number of pages	20
Journal	Econometrics Journal
Volume	27
Issue number	3
DOIs	https://doi.org/10.1093/ectj/utae012
Publication status	Published - 1 Sept 2024

JEL classifications

c32 - "Multiple or Simultaneous Equation Models: Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models"

Keywords

Local projections
impulse response analysis
high-dimensional data
honest inference
lasso
MACROECONOMICS
SELECTION

Access to Document

10.1093/ectj/utae012

1 Preprint

Local Projection Inference in High Dimensions
Adamek, R., Smeekes, S. & Wilms, I., 8 Sept 2022, Cornell University - arXiv, 31 p. (arXiv.org; No. 2209.03218).
Research output: Working paper / Preprint › Preprint

Open Access
File
119 Downloads (Pure)

Cite this

@article{9b19db1888344d019de08e4dbc57aaa8,

title = "Local projection inference in high dimensions",

abstract = "In this paper, we estimate impulse responses by local projections in high-dimensional settings. We use the desparsified (de-biased) lasso to estimate the high-dimensional local projections, while leaving the impulse response parameter of interest unpenalized. We establish the uniform asymptotic normality of the proposed estimator under general conditions. Finally, we demonstrate small sample performance through a simulation study and consider two canonical applications in macroeconomic research on monetary policy and government spending.",

keywords = "Local projections, impulse response analysis, high-dimensional data, honest inference, lasso, MACROECONOMICS, SELECTION",

author = "Robert Adamek and Stephan Smeekes and Ines Wilms",

note = "data source: 1) McCracken, M.W. & Ng, S. (2015). FRED-MD: A Monthly Database for Macroeconomic Research. Version 2021-08. Retrieved from https://research.stlouisfed.org/econ/mccracken/fred-databases/ 2) Ramey, VA. & Zubairy, S (2018). Government Spending Multipliers in Good Times and in Bad: Evidence from U.S. Historical Data. Dataset retrieved from https://econweb.ucsd.edu/~vramey/research.html",

year = "2024",

month = sep,

day = "1",

doi = "10.1093/ectj/utae012",

language = "English",

volume = "27",

pages = "323--342",

journal = "Econometrics Journal",

issn = "1368-4221",

publisher = "Oxford University Press",

number = "3",

}

TY - JOUR

T1 - Local projection inference in high dimensions

AU - Adamek, Robert

AU - Smeekes, Stephan

AU - Wilms, Ines

N1 - data source: 1) McCracken, M.W. & Ng, S. (2015). FRED-MD: A Monthly Database for Macroeconomic Research. Version 2021-08. Retrieved from https://research.stlouisfed.org/econ/mccracken/fred-databases/ 2) Ramey, VA. & Zubairy, S (2018). Government Spending Multipliers in Good Times and in Bad: Evidence from U.S. Historical Data. Dataset retrieved from https://econweb.ucsd.edu/~vramey/research.html

PY - 2024/9/1

Y1 - 2024/9/1

N2 - In this paper, we estimate impulse responses by local projections in high-dimensional settings. We use the desparsified (de-biased) lasso to estimate the high-dimensional local projections, while leaving the impulse response parameter of interest unpenalized. We establish the uniform asymptotic normality of the proposed estimator under general conditions. Finally, we demonstrate small sample performance through a simulation study and consider two canonical applications in macroeconomic research on monetary policy and government spending.

AB - In this paper, we estimate impulse responses by local projections in high-dimensional settings. We use the desparsified (de-biased) lasso to estimate the high-dimensional local projections, while leaving the impulse response parameter of interest unpenalized. We establish the uniform asymptotic normality of the proposed estimator under general conditions. Finally, we demonstrate small sample performance through a simulation study and consider two canonical applications in macroeconomic research on monetary policy and government spending.

KW - Local projections

KW - impulse response analysis

KW - high-dimensional data

KW - honest inference

KW - lasso

KW - MACROECONOMICS

KW - SELECTION

U2 - 10.1093/ectj/utae012

DO - 10.1093/ectj/utae012

M3 - Article

SN - 1368-4221

VL - 27

SP - 323

EP - 342

JO - Econometrics Journal

JF - Econometrics Journal

IS - 3

ER -

Local projection inference in high dimensions

Abstract

JEL classifications

Keywords

Access to Document

Fingerprint

Research output

Local Projection Inference in High Dimensions

Cite this