Abstract
In a recent interesting article, Minier (Citation2003) uses results from regression tree techniques to conclude that, for countries with high stock market capitalization, there exists a significantly positive relationship between the level of financial development and economic growth. However, the relationship becomes significantly negative for low-capitalization subsample. Using same data set, this short article re-examines Minier's results by the threshold regression approach of Hansen (Citation2000), but finds no evidence supporting that the full sample can be classified into two distinct regimes as in Minier (Citation2003). Our formal test result suggests that one should be more cautious regarding the conclusions made by Minier (Citation2003).
1 The data and computer code (with output) used in this article are available upon request from the corresponding author.
Acknowledgements
We are grateful to Ross Levine and Sara Zervos for making data available and Bruce E. Hansen for sharing the GAUSS code. All remaining errors are our own responsibility.
Notes
1 The data and computer code (with output) used in this article are available upon request from the corresponding author.