Managed float exhange rate system and the efficiency of Thailand foreign exchange market

Abstract

This thesis uses the Cointegration technique and Error Correction Model by Engle and Granger (1987), Correlogram and Q-statistic to test the foreign exchange market efficiency hypothesis(MEH) in the early managed float exchange rate system in Thailand. Moreover, this thesis attempts to find the answer whether the failure of the MEH is due to the risk premium 1 expectation errors 1 or both. By using daily spot and one-month forward exchange rate of the Thai Baht per US. Dollar during the period July 2,1997 through February 23 1 1998 (169 observations) obtained from the Data stream, this thesis found that both spot and one-month forward exchange rate are non-statienary at the levels but stationary after taking the first differences. The results from the cointegration test show they have no long term relationship and along with the test results from Correlogram and Q-statistic, show significant serial correlation in the residuals. All the results imply that the Thailand foreign exchange market is inefficient. The result from another two cointergration tests, conclude that the violation of the market efficiency hypothesis in Thailand is due to the rejection of the ' no risk premium’ hypothesis, and thus the MEH is rejected.