Cookies?
Library Header Image
LSE Research Online LSE Library Services

Agnostic structural disturbances (ASDs): detecting and reducing misspecification in empirical macroeconomic models

Den Haan, Wouter J. ORCID: 0000-0001-6214-8156 and Drechsel, Thomas (2020) Agnostic structural disturbances (ASDs): detecting and reducing misspecification in empirical macroeconomic models. Journal of Monetary Economics. pp. 1-26. ISSN 0304-3932

[img] Text (Den-Haan_agnostic-structural-disturbances--published) - Published Version
Available under License Creative Commons Attribution.

Download (1MB)

Identification Number: 10.1016/j.jmoneco.2020.01.005

Abstract

Constructing empirical specifications for structural economic models is difficult, if not impossible. As shown in this paper, even minor misspecifications may lead to large distortions for parameter estimates and implied model properties. We propose a novel concept, namely an agnostic structural disturbance (ASD), that can be used to both detect and correct for misspecification of structural disturbances and is easy to implement. While agnostic in nature, the estimated coefficients and associated impulse response functions of the ASDs allow us to give them an economic interpretation. We adopt the methodology to the Smets–Wouters model and formulate an improved risk-premium and an improved investment-specific productivity disturbance.

Item Type: Article
Official URL: https://www.journals.elsevier.com/journal-of-monet...
Additional Information: © 2020 The Authors
Divisions: Economics
Subjects: H Social Sciences > HB Economic Theory
JEL classification: C - Mathematical and Quantitative Methods > C1 - Econometric and Statistical Methods: General > C13 - Estimation
C - Mathematical and Quantitative Methods > C5 - Econometric Modeling > C52 - Model Evaluation and Selection
E - Macroeconomics and Monetary Economics > E3 - Prices, Business Fluctuations, and Cycles > E30 - General
Date Deposited: 22 Jan 2020 09:54
Last Modified: 08 Nov 2024 20:12
URI: http://eprints.lse.ac.uk/id/eprint/103147

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics