A functional coefficient model view of the saving-investment relation

Two of the most discussed issues triggered by the so called Feldstein-Horioka puzzle in international macroeconomics are: What does the saving-investment (SI) relation really measure and how should the SI relation be measured? The first contribution of this study is to develop a new variant of functional coefficient models to analyse a wide range of determinants of the SI relation. Macroeconomic state variables such as openness, the age dependency ratio, government current and consumption expenditures are found to affect the SI relation significantly in the long run. As the second contribution, a new, factor based bootstrap approach is proposed for inferential issues in functional coefficient models. This approach can cope with heterogeneous error distributions and is proven to hold asymptotically. In simulation studies factor based bootstrap inference outperforms the wild bootstrap and pairs bootstrap approach according to its size features. Regarding current account imbalances as saving minus investment, the third contribution of the thesis is to test the stationarity of the current account imbalance by taking account of its `bounded' nature. Time-series as the current account imbalance to GDP are bounded not only by construction but also via potential policy controls or economic crises. Taking account of these bounds weakens the evidence against the prevalence of stochastic trends. At the aggregate level the null hypothesis of bounded integration cannot be rejected for the current account imbalances.