| Article ID: | iaor200837 |
| Country: | United Kingdom |
| Volume: | 23 |
| Issue: | 3 |
| Start Page Number: | 173 |
| End Page Number: | 196 |
| Publication Date: | Apr 2004 |
| Journal: | International Journal of Forecasting |
| Authors: | Camacho Maximo |
| Keywords: | financial, statistics: regression, government |
In this paper, I extend to a multiple-equation context the linearity, model selection and model adequacy tests recently proposed for univariate smooth transition regression models. Using this result, I examine the nonlinear forecasting power of the Conference Board composite index of leading indicators to predict both output growth and the business-cycle phases of the US economy in real time.