Estimation and Forecasting of the Dynamic Relationship between Stock Prices and Exchange Rates: A Comparative Study

This thesis examines the dynamic relationship between stock prices and exchange rates. Both long-run and short-run relationships between these variables are explored. The study uses daily time series data from China, the European Union, the United Kingdom and the United States. The period of study was divided into in-sample and out-of-sample data. The in-sample data set covered the period starting from January 3, 2000 to December 31, 2010 and included 22,976 observations after adjustments, whereas the outof-sample data extended from January 3, 2011 to March 31, 2015 and incorporated 8,848 observations after adjustments. The study uses the in-sample-data to apply cointegration tests, the Vector Auto Regression model (VAR), the Vector Error Correction Model, and Granger causality tests to examine the short and long-run relationship between stock prices and exchange rates in the countries of the sample. The results revealed a long-run relationship between stock prices and exchange rates running from the Euro Exchange Rate to the FTSE Eurotop 100 Index, which supports the Flow-Oriented Theory. Furthermore, this study showed that there is a unidirectional Granger-causality relationship running from the Chinese Exchange Rate to the Shanghai Stock Exchange Composite Index closing price in the short-run. This result supports the arguments of the Flow-Oriented Theory. Moreover, this study demonstrates that there is a unidirectional Granger-causality relationship running from the Dow Jones Industrial Average Index closing price to the US Exchange Rate in the short-run, which corresponds with the arguments of the Share-Oriented Theory. With regards to the United Kingdom, bidirectional causality has been found between the FTSE 100 Index closing price and the UK Exchange Rate in the short-run, which supports Flow-Oriented and Share-Oriented Theories. Furthermore, this study uses the out-of-sample data to apply a VAR Forecast for China, the United Kingdom and the United States because there is short-run relationship between stock prices and exchange rates. The out-of-sample data is also used to estimate the VECM Forecast for the European Union