This thesis consists of four chapters concerning the topics of nonlinearities and time scales in economics. The focus is on market frictions and price rigidities that may cause nonlinearities and different relationships between economic variables over time. It also focuses on applying robust econometrical methods. Chapter two evaluates the size and power of some nonlinear tests for panel unit roots in the presence of cross-sectional dependency and spatial dependency. Based on the simulated results some robust tests for nonlinear panel unit roots have been found.

Chapter three applies robust linear and nonlinear tests for panel unit roots in order to investigate the purchasing power parity theory in developing regions. The main finding is that nonlinearities is an important phenomenon in the real effective exchange rates in developing regions and that support for several regions may be found by applying the nonlinear panel unit root test. Chapter four investigates if positive asymmetric price transmission (APT) exists in the Swedish mortgage loan market. Here robust quantile regression is used on data for the Swedish SEB bank. The main contribution is that positive APT effects are found, which implies that there is a higher propensity for the bank to rapidly increase its mortgage interest rates for customers following an increase in its borrowing costs, compared with the propensity for the bank to decrease its customers’ mortgage rates subsequent to a corresponding borrowing cost decrease.

Chapter five investigates the causal relations between exchange rates and interest rate differentials using wavelets. Also the sign of the relationship between the two variables is studied using impulse response functions. The data used is for seven country pairs in which Sweden is included in all of the different combinations. In this chapter one key empirical finding is that the causal relationship between the two variables becomes stronger as the time scale increases. The other key empirical finding is that more evidence of negative relationships is found at the shorter time scales and more positive relationships at the longer time scales.