Essays on the impact of the Asian crisis on exchange rates

In the past two decades, there has been an extensive study on exchange rate economics. But research on the exchange rate related to currency crisis still remains as a challenging area. In this thesis, on the empirical side, we try to find the impact of 1997 Southeast Asian currency and financial crisis on both the nominal and real exchange rates in some east economies.;In the first chapter, we explore the validity of the Purchasing Power Parity (PPP) in seven economies that were affected by the 1997 Asian financial crisis. We develop a new panel structural break test and identify the common beginning and ending dates of the crisis. We apply the traditional unit root tests, the ADF and LLC statistics, and the latest ones, the CADF and CIPS statistics (Pesaran, 2007) to examine the null hypothesis of the non-stationarity in real exchange rates. The overwhelming rejection of the null hypothesis from the empirical evidence shows that PPP does not hold for most of the sample economies during the pre- and post-crisis period. However, for Indonesia and Taiwan, the null hypothesis of unit root is rejected during the period before the occurrence of the crisis. It provides us some evidence in support of the possibility that the occurrence of the financial crisis may affect the validity of PPP afterwards.;In the second chapter, we define contagion as a significant increase in spillover effects during the crisis period, after controlling for common international shocks. We apply a factor-structural VAR framework that permits a decomposition of forecast variance into three sources: common shocks, idiosyncratic shocks and spillovers effects. Using daily exchange rates and stock prices in nine Asian countries and regions, we present a comprehensive analysis of cross-market, cross-country transmission of shocks. We find that it is common shocks and idiosyncratic shocks that account for the most part of forecast variance in percentage change of daily asset prices. Based on our definition of contagion, there is no substantial contagion occurring during the Asian financial crisis.;In the last chapter, we investigate the empirical performance of the stochastic volatility (SV) model, when it is used to predict the weekly nominal exchange rate for the sample Asian economies during the Asian financial crisis. The stochastic volatility model is estimated by Markov Chain Monte Carlo method. We compare the forecasting performance of the stochastic volatility model to several benchmark models and find that the stochastic volatility model indeed outperforms the random walk model and the GARCH models, when the volatility of the nominal exchange rates is relatively high, i.e. during the crisis period.