Based On Co-integration, VEC And GARCH-M Approach To Testing The Relationship Of RMB/USD Spot And Forward Exchange Rates

Exchange rate is one of the most important factors in open macroeconomic. It is necessary for economic development to have a suitable exchange rate and an efficient foreign exchange market. One view of market efficiency states that the currency prices reflect all available information. When this is applied to the foreign exchange market, it implies that economic agent's expectations about future values of exchange rate determinants are fully reflected in the forward rates. In a no-arbitrage environment, the both assumptions of risk neutrality and rational expectations imply that the forward foreign exchange rate should be an unbiased predictor of the corresponding spot rate.July 21st, 2005 is a meaningful day for China foreign exchange history, because Reform of Exchange Rate Regime is announced that day. China initiated its new exchange rate regime, which is a controlled floating one, based on market supply and demand, while adjusted by a basket of currencies. It is becoming an urgent topic to study and explore the future development of Chinese foreign exchange derivatives market.This paper is focusing in three interesting problems.Firstly, I estimate RMB/USD exchange rate based on the UIP and revised UIP model with transaction costs. The difference of the real RMB/USD forward exchange rates and the estimating ones is very large using above the two methods. And with the period is longer, the difference is larger. The results show that the RMB/USD exchange rate was appreciated before July, 2004 and was depreciated after that time. Secondly, I investigate that the spot exchange rates and forward exchange rates appear to be cointegrated in the long run. This paper established the VCM models for the different period RMB/USD forward exchange rates and the subsequent spot exchange rates. I have a result that the adjusting speed of the shorter period forwards is quicker than the longer period forwards. By the Granger testing, I have a conclusion that the spot exchange rate is the Granger cause of the different kinds of forward exchange rates in the long run. But in the short run , neither is the case.Finally, I modeled the conditional covariance as a GARCH and GARCH-in-mean process. I apply the models to a sample of RMB/USD in China foreign exchange market. The forward discount has a positive coefficient in the 7days and 20 days forward exchange rate models, the forward discount at least points in the right direction. But In the 30 days forward model, the forward discount points in the opposite direction. To us this suggests that a time-varying exchange risk premium may be the explanation for traditional findings of bias. It suggests that on the average an investor can earn positive profits by investing in the U.S. dollar when the U.S. dollar has a forward discount with regard to the RMB, and investing in the RMB when the U.S. dollar commands a forward premium.Estimation results suggest that the forward rate bias exists in the China foreign exchanger market. The forward premium and the expected change in the spot exchange rate move with positive or passive direction. This shows that we could not predict the future spot exchange rate in the base of the forwards exchange rate. In this sense, China foreign exchange rate market is not efficient.There is much future research about those discussing problems above. For example, whether establishing more general model including the jump point on 21st,July, 2005. And we test the VCE model's forecasting power, and so on.