Trend model testing of growth convergence in 15 OECD countries, 1946-1997

A framework based on a linear deterministic trend function is introduced in order to model growth convergence. The approach is a practical solution to the nonlinearity and nonstationarity found in the convergence of output-per-capita gaps between the USA and 14 OECD countries in 1946-1997. Convergen...

Full description

Saved in:
Bibliographic Details
Published inApplied economics Vol. 34; no. 2; pp. 133 - 142
Main Author Linden, Mikael
Format Journal Article
LanguageEnglish
Published Taylor & Francis Group 01.01.2002
Taylor and Francis Journals
SeriesApplied Economics
Subjects
Applied economics
Convergence
Data analysis
Economic models
Economics
History
Japan
Model testing
OECD
Online AccessGet full text

Cover

More Information
Summary:A framework based on a linear deterministic trend function is introduced in order to model growth convergence. The approach is a practical solution to the nonlinearity and nonstationarity found in the convergence of output-per-capita gaps between the USA and 14 OECD countries in 1946-1997. Convergence is found to be a typical feature of European OECD countries and Japan, but for some major countries it altered since the beginning of 1980. Some evidence of divergence exists too. Valid statistical inference on trend-growth estimates is based on sample-size standardized t - and Wald-test statistics, since the model residuals are close to I(1). Small-sample test values are derived using bootstrap methods.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:0003-6846
1466-4283
DOI:10.1080/00036840010025119