6. Conclusions

We use macroeconomic panel data from 16 countries to estimate a model that enables us to find evidence for the UIP. Interestingly enough, we find strong evidence for UIP when we use long-term interest rate differentials. Moreover, the corresponding short-term interest rate coefficient turns out to be zero. We conclude that UIP does not hold in the short-run. Since the static fixed effects model may be inefficient and biased because of significant first order autocorrelation, we specify a dynamic panel data model, which is in turn estimated by GMM. Results from the dynamic model are in line with what has been found before from the static model.

Method and Data Technologies   MD*TECH Method and Data Technologies
  http://www.mdtech.de  mdtech@mdtech.de