Local Projection Inference in High Dimensions

Robert Adamek; Stephan Smeekes; Ines Wilms

Local Projection Inference in High Dimensions

Robert Adamek, Stephan Smeekes, Ines Wilms

Research output: Working paper / Preprint › Preprint

38 Downloads (Pure)

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
Publisher	Cornell University - arXiv
Number of pages	31
Publication status	Published - 8 Sept 2022

Publication series

Series	arXiv.org
Number	2209.03218
ISSN	2331-8422

JEL classifications

c22 - "Single Equation Models; Single Variables: Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models"
c12 - Hypothesis Testing: General

Keywords

high-dimensional data
honest inference
impuls responses
lasso
time series

Access to Document

pdfSubmitted manuscript, 684 KBLicence: CC BY

https://arxiv.org/pdf/2209.03218.pdfLicence: CC BY

Cite this

@techreport{126c2b6a391c47a1bfe8ca49e4c50b90,

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 = "high-dimensional data, honest inference, impuls responses, lasso, time series",

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 = "2022",

month = sep,

day = "8",

language = "English",

series = "arXiv.org",

number = "2209.03218",

publisher = "Cornell University - arXiv",

address = "United States",

type = "WorkingPaper",

institution = "Cornell University - arXiv",

}

TY - UNPB

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 - 2022/9/8

Y1 - 2022/9/8

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 - high-dimensional data

KW - honest inference

KW - impuls responses

KW - lasso

KW - time series

M3 - Preprint

T3 - arXiv.org

BT - Local Projection Inference in High Dimensions

PB - Cornell University - arXiv

ER -