The yield curve as a predictor of future economic activity
(2012) BUSP69 20121Department of Business Administration
 Abstract
 The primary purpose of this study is to determine whether the yield curve has the ability to predict real future economic activity in sixteen separate countries. The secondary purpose is to determine whether adding nonmonetary variables to the model increases the predictive power of the forecast model. As a first step, basic OLS regressions are conducted. Both the yield spreads and the yield spread combined with nonmonetary variables ability to predict future real GDP growth is measured. As a second approach, probit models are estimated measuring the ability of the yield spread to forecast the probability of a future above trend real GDP. Forecast horizons ranging from one to twenty quarters ahead are used. The study finds that the yield... (More)
 The primary purpose of this study is to determine whether the yield curve has the ability to predict real future economic activity in sixteen separate countries. The secondary purpose is to determine whether adding nonmonetary variables to the model increases the predictive power of the forecast model. As a first step, basic OLS regressions are conducted. Both the yield spreads and the yield spread combined with nonmonetary variables ability to predict future real GDP growth is measured. As a second approach, probit models are estimated measuring the ability of the yield spread to forecast the probability of a future above trend real GDP. Forecast horizons ranging from one to twenty quarters ahead are used. The study finds that the yield curve holds strong significant predictive power of forecasting future GDP growth in the cases of Australia, Norway, Poland, Switzerland, United Kingdom, and USA. For all these countries, the yield spread is found to be a significant predictor of future real GDP growth during at least six of the eleven investigated forecast horizons. In all cases, an increase of the yield spread is found to predict an increase in real GDP growth. It is also found that the models may be enhanced by adding the unemployment rate and stock index variables as independent variables. Although these variables, in many cases show significant predictive power, including them in the model should be done with caution since they often only provide a negligible increase of the explanatory power (as measured by adjusted R2) of the models. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/studentpapers/record/2837252
 author
 Wickström, Hans ^{LU} and Smith, Axel ^{LU}
 supervisor

 Göran Anderson ^{LU}
 organization
 course
 BUSP69 20121
 year
 2012
 type
 H2  Master's Degree (Two Years)
 subject
 keywords
 Yield curve, yield spread, OLSregression, probit regression, nonmonetary variable, unemployment, car sales, stock index, monetary policy, cyclical GDP, real GDP growth
 language
 English
 id
 2837252
 date added to LUP
 20120627 13:57:46
 date last changed
 20120627 13:57:46
@misc{2837252, abstract = {The primary purpose of this study is to determine whether the yield curve has the ability to predict real future economic activity in sixteen separate countries. The secondary purpose is to determine whether adding nonmonetary variables to the model increases the predictive power of the forecast model. As a first step, basic OLS regressions are conducted. Both the yield spreads and the yield spread combined with nonmonetary variables ability to predict future real GDP growth is measured. As a second approach, probit models are estimated measuring the ability of the yield spread to forecast the probability of a future above trend real GDP. Forecast horizons ranging from one to twenty quarters ahead are used. The study finds that the yield curve holds strong significant predictive power of forecasting future GDP growth in the cases of Australia, Norway, Poland, Switzerland, United Kingdom, and USA. For all these countries, the yield spread is found to be a significant predictor of future real GDP growth during at least six of the eleven investigated forecast horizons. In all cases, an increase of the yield spread is found to predict an increase in real GDP growth. It is also found that the models may be enhanced by adding the unemployment rate and stock index variables as independent variables. Although these variables, in many cases show significant predictive power, including them in the model should be done with caution since they often only provide a negligible increase of the explanatory power (as measured by adjusted R2) of the models.}, author = {Wickström, Hans and Smith, Axel}, keyword = {Yield curve,yield spread,OLSregression,probit regression,nonmonetary variable,unemployment,car sales,stock index,monetary policy,cyclical GDP,real GDP growth}, language = {eng}, note = {Student Paper}, title = {The yield curve as a predictor of future economic activity}, year = {2012}, }