Estimating the effect of a variable in a high-dimensional linear model

Peter Sandholt Jensen, Allan Würtz

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

Resumé

A problem encountered in some empirical research, e.g. growth empirics, is that the potential number of explanatory variables is large compared to the number of observations. This makes it infeasible to condition on all variables in order to determine whether a variable of interest has an effect. We assume that the effect is identified in a high‐dimensional linear model specified by unconditional moment restrictions. We propose a new method that provides a consistent estimator of the effect when the variable of interest is conditional mean independent of excluded variables. Existing methods are consistent when excluded variables do not explain the outcome, but not under the conditional mean independence assumption. We also demonstrate that the new method has good properties in a Monte Carlo study.
OriginalsprogEngelsk
TidsskriftEconometrics Journal
Vol/bind15
Udgave nummer2
Sider (fra-til)325-357
ISSN1368-4221
DOI
StatusUdgivet - 2012

Fingeraftryk

Empirics
Monte Carlo study
Estimator
Empirical research

Citer dette

@article{377ea71808e14f809c984b94a80b3a0e,
title = "Estimating the effect of a variable in a high-dimensional linear model",
abstract = "A problem encountered in some empirical research, e.g. growth empirics, is that the potential number of explanatory variables is large compared to the number of observations. This makes it infeasible to condition on all variables in order to determine whether a variable of interest has an effect. We assume that the effect is identified in a high‐dimensional linear model specified by unconditional moment restrictions. We propose a new method that provides a consistent estimator of the effect when the variable of interest is conditional mean independent of excluded variables. Existing methods are consistent when excluded variables do not explain the outcome, but not under the conditional mean independence assumption. We also demonstrate that the new method has good properties in a Monte Carlo study.",
author = "Jensen, {Peter Sandholt} and Allan W{\"u}rtz",
year = "2012",
doi = "10.1111/j.1368-423X.2011.00362.x",
language = "English",
volume = "15",
pages = "325--357",
journal = "Econometrics Journal",
issn = "1368-4221",
publisher = "Heinemann",
number = "2",

}

Estimating the effect of a variable in a high-dimensional linear model. / Jensen, Peter Sandholt; Würtz, Allan.

I: Econometrics Journal, Bind 15, Nr. 2, 2012, s. 325-357.

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

TY - JOUR

T1 - Estimating the effect of a variable in a high-dimensional linear model

AU - Jensen, Peter Sandholt

AU - Würtz, Allan

PY - 2012

Y1 - 2012

N2 - A problem encountered in some empirical research, e.g. growth empirics, is that the potential number of explanatory variables is large compared to the number of observations. This makes it infeasible to condition on all variables in order to determine whether a variable of interest has an effect. We assume that the effect is identified in a high‐dimensional linear model specified by unconditional moment restrictions. We propose a new method that provides a consistent estimator of the effect when the variable of interest is conditional mean independent of excluded variables. Existing methods are consistent when excluded variables do not explain the outcome, but not under the conditional mean independence assumption. We also demonstrate that the new method has good properties in a Monte Carlo study.

AB - A problem encountered in some empirical research, e.g. growth empirics, is that the potential number of explanatory variables is large compared to the number of observations. This makes it infeasible to condition on all variables in order to determine whether a variable of interest has an effect. We assume that the effect is identified in a high‐dimensional linear model specified by unconditional moment restrictions. We propose a new method that provides a consistent estimator of the effect when the variable of interest is conditional mean independent of excluded variables. Existing methods are consistent when excluded variables do not explain the outcome, but not under the conditional mean independence assumption. We also demonstrate that the new method has good properties in a Monte Carlo study.

U2 - 10.1111/j.1368-423X.2011.00362.x

DO - 10.1111/j.1368-423X.2011.00362.x

M3 - Journal article

VL - 15

SP - 325

EP - 357

JO - Econometrics Journal

JF - Econometrics Journal

SN - 1368-4221

IS - 2

ER -