Testing the null of a low dimensional growth model

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

Resumé

Udgivelsesdato: Februar
OriginalsprogEngelsk
TidsskriftEmpirical Economics
Vol/bind38
Udgave nummer1
Sider (fra-til)193-215
ISSN0377-7332
DOI
StatusUdgivet - 1. feb. 2010

Fingeraftryk

Growth Model
Null
Testing
Averaging
High-dimensional
Monte Carlo Simulation
simulation
Growth model
Alternatives
Estimate
evidence
Observation

Citer dette

@article{5e03b7201ab811deb280000ea68e967b,
title = "Testing the null of a low dimensional growth model",
abstract = "Applied growth researchers face a sample with more regressors than observations when using cross-country data. In this paper, we examine simple procedures for testing if a true submodel has been found. Specifically, we examine three tests that can be used to test a low dimensional growth model against a high dimensional alternative, when there are more regressors than observations. We examine the finite sample properties of the tests using Monte Carlo simulations. We apply the tests to a specific growth dataset that has more variables than observations and find evidence against the null that no variables are important. Finally, we apply Bayesian Averaging of Classical Estimates and reach a similar conclusion.",
author = "Jensen, {Peter Sandholt}",
year = "2010",
month = "2",
day = "1",
doi = "10.1007/s00181-009-0262-1",
language = "English",
volume = "38",
pages = "193--215",
journal = "Empirical Economics",
issn = "0377-7332",
publisher = "Heinemann",
number = "1",

}

Testing the null of a low dimensional growth model. / Jensen, Peter Sandholt.

I: Empirical Economics, Bind 38, Nr. 1, 01.02.2010, s. 193-215.

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

TY - JOUR

T1 - Testing the null of a low dimensional growth model

AU - Jensen, Peter Sandholt

PY - 2010/2/1

Y1 - 2010/2/1

N2 - Applied growth researchers face a sample with more regressors than observations when using cross-country data. In this paper, we examine simple procedures for testing if a true submodel has been found. Specifically, we examine three tests that can be used to test a low dimensional growth model against a high dimensional alternative, when there are more regressors than observations. We examine the finite sample properties of the tests using Monte Carlo simulations. We apply the tests to a specific growth dataset that has more variables than observations and find evidence against the null that no variables are important. Finally, we apply Bayesian Averaging of Classical Estimates and reach a similar conclusion.

AB - Applied growth researchers face a sample with more regressors than observations when using cross-country data. In this paper, we examine simple procedures for testing if a true submodel has been found. Specifically, we examine three tests that can be used to test a low dimensional growth model against a high dimensional alternative, when there are more regressors than observations. We examine the finite sample properties of the tests using Monte Carlo simulations. We apply the tests to a specific growth dataset that has more variables than observations and find evidence against the null that no variables are important. Finally, we apply Bayesian Averaging of Classical Estimates and reach a similar conclusion.

U2 - 10.1007/s00181-009-0262-1

DO - 10.1007/s00181-009-0262-1

M3 - Journal article

VL - 38

SP - 193

EP - 215

JO - Empirical Economics

JF - Empirical Economics

SN - 0377-7332

IS - 1

ER -