This doctoral dissertation is a collection of six articles. Special attention is paid to model specification and the time-series properties of the data applied in each case. The first two articles deal with fiscal policy in Sweden and the effects of EMU criteria convergence. Besides considering the integral and cointegral properties of the macroaggregates, fixed coefficients as well as random coefficients models are used to test the underlying hypotheses. The third article investigates the import demand function using the Johansen and Juselius procedure. The approach applied here is claimed to reveal the pertinent demand elasticities with more accuracy and reliability than other methods. In the fourth article, the long-run convergence between the trade balance and the terms of trade is explored. The fifth article investigates Balassa's export-led growth hypothesis for a panel of five countries by constructing VAR models. The Granger causality procedure, developed by Toda and Yamamoto, is used to improve the standard F-statistics in the causality test process. The objective of the last article is to test for informational efficiency of the Korean stock market by the bootstrap simulation technique.