本研究主要評估雙變量誤差修正模型、基因演化模式及倒傳遞類神經網路在價格預測的預測效果以找出最適的預測模型，及探討1987年及1989年美國股市崩盤前後國際貨幣市場間價格領先落後關係的變化。不但探討三個月期歐洲美元(外在)市場和三個月期美國國庫券(國內)市場彼此間的互動關係，同時利用傳統時間序列模型與人工智慧模型來預測期貨價格的變動與方向的預測並探討兩種模型的優劣。本研究以Engle and Granger (1987)所提出的共整(cointegration)模式，來檢定三個月期美國國庫券與三個月期歐洲美元利率期貨市場價格是否有長期均衡關係。並且使用誤差修正搭配EGARCH模式分別與遺傳基因工程(Genetic Algorithm)與倒傳類神經網路(Back Propagation network)結合來做價格變動的預測，探討傳統模型結合遺傳基因工程與倒傳類神經網路是否優於單獨地傳統時間序列模型。
　　然而，由於資金在國際貨幣市場間的移動障礙日趨減少且資訊傳送迅速，一國貨幣市場所受到任何消息面的影響很快地透過市場參與者套利交易(Arbitrage)活動，造成資金在彼此間移動，所以加速市場間的互動關係。故此本研究利用日內資料以有效掌握彼此市場間對訊息的反應程度�即以每十分鐘的資料作下列之比較與分析，實證結果如下：
(1) 以Engle and Granger共整合模式檢定三個月期美國國庫券與三個月期歐洲美元利率期貨價格發現�無論崩盤前後均具有一長期均衡的關係，表示美國國庫券與歐洲美元利率期貨價格在短期下可能受新訊息(innovation)的影響偏離均衡關係，但在套利者的參與，長期而言仍然維持價格趨勢的一致性。
(2) EGARCH配適誤差修正模型(ECM)中之條件平均數顯示�根據十分鐘的資料發現在1987年股市崩盤後，三個月期美國國庫券期貨價格與三個月期歐洲美元期貨價格間具有明顯的領先落後關係，且存在回饋(Feedback)效果。即兩市場是呈現相互領先落後的關係，所以國內貨幣市場或外在貨幣市場的資訊均會傳遞到另一市場。
(3) 無論在預測價格的精確性或預測價格變動方向的準確性來說，人工智慧皆優於傳統的時間序列模型。
(4) 進一步以Wilcoxon sign rank sum test得到顯著的結果，即無論在崩盤前或崩盤後，遺傳基因在預測的精確性上皆優於倒傳遞類神經網路及雙變量誤差修正模型。而Fisher exact test的結果顯示，無論在崩盤前或崩盤後，遺傳基因工程在預測方向的準確上為最佳，倒傳遞類神經網路次之，而雙變量誤差修正模型為最差。

　　This paper examines mainly that the linkages　and　predictive power between the Treasury bill and Eurodollar futures markets applying the data of pre- and post-American stock markets crashed on October in 1987 and 1989. The Error-Correction Model (ECM) obtained from bivariate EC-EGARCH Modeling techniques are utilized to examine the long-term equilibrium relation, short-term lead-lag relationship between the TB and ED futures markets. Early researchers have proved that ECM is better than Granger Causality (Lin & Swanson, 1993). Also, ECM with innovations following a bivariate EGARCH (Nelson, 1991) process is used to describe the joint distributions of TB future prices and ED futures prices, which can explore completely the dynamics　interactions between the two markets, since early　researcher ignored that the conditional variance in each market is an asymmetric function of its own and other related market.
　　Rauscher (1997) proved that back propagation network is better than ECM model in directional predictions of　exchange changes applying monthly data of Mark to Dollar in 1986-1997. Therefore, this study will combines error correction model with Genetic Algorithm and back　propagation network (call “the integrated model”) respectively to predict the Treasury bill and Eurodollar futures price change to see if the integrated model outperforms the model using only error correction model.
　　However, with the reduction in barriers to　international capital flows, arbitragers invest their funds in the money market as soon as they get any information in the market. For this reason, the relationship between the two money markets, the three-month TB and the three-month ED, would get more closely than before. So we adopt short trading time interval of intraday data. In this paper the intraday data of 10-minute are utilized instead of long trading time interval.
　　The results are as below:
1. The results of DF and ADF tests for cointegration show that there exists a long-run equilibrium between the U.S. Treasury Bill and Eurodollar futures prices in both pre-and post-crash periods. The results indicate that if TB and ED futures prices deviate away from the equilibrium　relationship in the short-term period, which is due to the innovations of respective market, the arbitrage will exist. For this reason, TB and ED keep the equilibrium price in the long-term period due to the arbitrage.
2. Based on 10-minute data, the results of the conditional mean equations in unrestricted EC-EGARCH Model indicate that the lead-lag relationships are statistically significant between U.S. Treasury Bill and Eurodollar futures prices in most of the periods. For this reason, we conclude that the feedback relationship exists between TB and ED money markets in most of the periods.
3. The results of predictive preciseness show that　overall the integrated models are better than error　correction model.
4. Genetic Algorithm model is the best in directional prediction. Back Propagation network is the second best and EC-EGARCH is the worst.
5. We further test the predictive effects using Wilcoxon sign rank sum test and Fisher exact test. The results show that three models are significantly different in most of the periods in both the preciseness of predictions and the accuracy of directional prediction.

Akaike, H. “Fitting Autoregressive Model for reduction”, Annual of Institute of Statistical Mathematics, 1969, January, pp.243-247.

Argy, V. and Zoran, H. “Financial Integration and Interest Rate Linkages in Industrial Countries, 1958-71”, International Monetary Fund Staff Papers, 1973, March, vol. 20, pp.1-77.

Arjun, C., Mukesh, C. and Rohan, C. D. “Price Discovery in Strategically Linked Markets: The TED Spread and its Constituents”, The Journal of Derivatives, 1999, Summer, pp.77-87.

Ball, C. A. and Torous, W. N. “ The Stochastic Volatility of Short-Term Interest Rates: Some International Evidence”, The Journal of Finance, 1999, December, vol.6, pp.2339-2359.

Booth, G. G. and Tse, Y. “Long Memory in Interest Rate Futures Markets: A Fractional Cointegration Analysis”, Journal of Futures Markets, 1995, vol.15, no.5, pp.573-584.

Cavanaugh, K. L. “Price Dynamics in Foreign Currency Futures Markets”, Journal of International Money and Finance, 1987, vol.6, pp.295-314.

Cheung, D. W. “The Impulse of Stock Market Volatility and the Market Crash of
October 1987”, Journal of Business Finance & Accounting, 2000, June/July, pp.761-775.

Dickey, D. A. and Fuller, W. A. “Distribution of the Estimates for Autoregressive Time Series With a Unit Root ”, Journal of the American Statistical Association, 1979, June, vol.24, pp.427-431.

Engle, R. F. and Granger, C. W. J. “Co-integration and Error Correction :
Representation , Estimation ,and Testing”, Econimetrica 1987, March, vol.55,
pp.251-76.

Fleming, M. J. and Remolona, E. M. “Price Formation and Liquidity in the U.S.
Treasury Market: The Response to Public Information, Journal of Finance, 1999,
vol.5, pp.1901-1915.

Fung, H. G. and Isbery, S. C. “The international transmission of Eurodollar and U.S. interest rates: A cointegration analysis”, Journal of Banking and Finance, 1992, vol.16, pp.757-769.

Geweke, J., Meese, R. and Dent, W. “Comparing Alternative Tests of Causality in Temporal Systems: Analytic Results and Experimental Evidence”, Journal of
Econometrics, 1983, February, vol.21, pp.161-194.

Giddy, I. H., Gunter, D. and Sangkee, M. “Interest Rates in the U.S. and Eurodollar Markets.” Weltwirtschaftliches Archiv, 1979, vol.115, pp.51-67.

Granger, C. W. J. “Investigating Causal Relations by Econometric Models and
Cross-Spectral Methods”, Econometrica, 1969, July, vol.37, pp.424-438.

Grudnitski, B. & L. Osburn,〝Forecasting S&P and Gold Futures Prices : An Application of Neural Networks,〞September 1993, pp.631-643.

Hartman, D. G. “The International Financial Market and U.S. Interest Rates”, Journal of International Money and Finance, 1984, April, vol.3, pp.91-103.

Hendershott, P. H. “The Structure of international Interest Rates: The U.S. Treasury Bill Rate and The Eurodollar Deposit Rate.” The Journal of Finance, 1967, vol. 22, pp.455-465.

Helmut, M. “ The Eurocurrency Market and the Autonomy of U.S. Monetary Policy”, Columbia Journal of World Business, 1979, Fall, pp.32-37

Hendershott, P. H. “The Structure of International Interest Rates: The U.S. Treasury Bill Rate and the Eurodollar Deposit Rate”, Journal of Finance, 1967, September, vol.22, pp.455-465.

Hugon, J. H. “Past and Future of the Eurodollar Market”, Financial Analysis Journal, 1971, September/October, vol.27, pp. 5-23.

Kaen, F. R. and Hachey, G. A. “Eurocurrency and National Money Market Interest Rates: An Empirical Investigation of Causality” Journal of Money, Credit and Banking, 1983, August, vol.15, pp.327-338.

Kanas, A. “Linkages between the U.S. and European Equity Markets: Further
Evidence from Cointegration Tests”, Applied Financial Economics, 1998, vol.8.

Kennedy, P. “A GUIDE To Econometrics”, Blackwell Publishers Ltd, 1998, Fourth Edition.

Koutmos, G. and Tucker, M. “Temporal Relationships and Dynamic Interactions
Between Spot and Futures Stock Markets” Journal of Futures Markets, 1996, vol.16, no.1, pp.55-68.

Krol, R. “The Term Structure of Eurodollar Interest Rates and its Relationship to the U.S. Treasury-Bill Market”, Journal of International Money and Finance, 1987, vol. 6, pp.339-354.

Kwack, S. Y. “The Structure of International Interest Rates: An extensin of
Hendershott’s Tests”, The Journal of Finance, 1971, pp.897-900.

Laopodis, N. T. “Currency Substitution and European Monetary Union”, Journal of Applied Business Research, 1998, vol.14, pp.74-51.

Levin, J. H. “The Eurodollar Market and the International Transmission of Interest Rates”, Canadian Journal of Economics, 1974, May, vol.7, pp.205-224.

Leung, M.T., A.S. Chen & H. Daouk,〝Forecasting exchange rates using general regression neural networks,〞Computers & Operations Research, 2000, pp.1093-1110.

Lin, A. and Swanson, P. E. “Measuring global Money market Interrelationships : Aninvestigation of five maior world currencies”, Journal of Banking and Finance, 1993, vol. 17, pp.609-628.

Maddala, G. S. “Introduction to Econometrics”, Prentice-Hall, Inc., 1992, 3rd Edition.

Macdonald, R. and Taylor, M. P. “The monetary model of the exchange rate: long-run relationships, short-run dynamics and how to beat a random walk”, Journal of International Money and Finance, 1994, vol.13, pp.276-290.

Makridakis, S., "Accuracy Measures: Theoretical and Practical Concerns", International Journal of Forecasting, Vol. 9, No. 4, 1993.

Malek, L. “The Role of TED Spread and Confidence Index in Explaining the
Behavior of Stock Prices”, American Business Revies, 2000, June, pp.9-11.

Michael, O. L. “A Note with Quantiles of the Asympototic Distribution of the
Maximum Likelihood Cointegration Rank Test Statistics”, Oxford Bulletin of
Economics and Statistics, 1992, vol.54, no.3, pp.461-473.

Nelson, D. B. “Conditional Heteroskedasticity in Asset Returns: A New Approach”, Econometrica, 1991, March, vol.59, no.2, pp.347-370.

Rauscher, F.A.,〝Exchange Rate Forecasting: a Neural VEC Approach to Non-Linear Time Series Analysis,〞Journal of Time Series Analysis, 1997, pp.461-471.

Ritchken, P. and Trevor, R. “ Pricing Options under Generalized GARCH and
Stochastic Volatility Processes”, Journal of Finance, 1999, February, vol.1,
pp.377-402.

Schnitzel, P. “Testing for the Direction of Causation between the
Domestic Monetary Base and the Eurodollar System”, Weltwirtschaftliches Archiv, 1983, vol.119, pp.616-629.

Susmll, R. “Hourly volatility spillovers between international equity markets, Journal of International Money and Finance, 1994, vol.13, pp.3-25.

Swanson, P. E. “Capital Market Integration over the Past Decade: The Case of the U.S. Dollar”, Journal of International Money and Finance 1987, vol.6, pp.215-225.

Tucker, A. L. “Financial Futures Options, & Swaps”, Info Access & Distribution Pte Ltd, 1992.

Wu, Chien-lang ,”The Research of Genetic Algorithm in Stock Market Prediction and Trading Strategy.” Thesis of Graduate School of Business Administration In NSYSU, 1995, June.

Wong, Chung-hao, “Compare the Prediction Effect of Cointegration Model with Artificial Neural Network”, Thesis of Graduate School of Economics In NCHU ,1993, June.

Yiuman, T and Geoffrey B., “Common Volatility and Volatility Spillovers between U.S. and Eurodollar Interest Rates: Evidence from the Futures Market” Journal of Economics and Business, 1996, vol.48, pp.299-312.

Yu, S. “A Study of Integration and Efficiency in Money Market-Based on the Three-Month Treasury Bill and Eurodollar Futures Prices” Thesis of Graduate School of Accounting In NCKU, 1995, June.