| Summary: | 碩士 === 國立中正大學 === 經濟系國際經濟學研究所 === 104 === Abstract
Since the official launch of TAIEX Futures (Taiwan Stock Exchange Capitalization Weighted Stock Index Futures) on July 21, 1998 and its trading until the end of 2015, the average daily volume as well as the number of accounts has been increasing. This fact shows that futures market has been warm. The investors are keen to pursue the excess returns.
This study is to investigate the design of the trading system to find out the strategy to excess returns, in order to apply it in the futures market. The object of the study is TAIEX Futures. The period of the study is from July 21, 1998 to December 31, 2015. The trading price of the futures is based on the sixty minutes cycle. The period is divided into seven sub-periods and one complete period. Fifteen combination of moving average strategy, twenty four combinations of the filter rule strategy, as well as sixteen combinations of my proposed strategy will be applied for analysis and to compare their potential of gaining profit. In doing so, it is hoped to find out the best strategy for future application. The result of this study is as follows:
1. After the examination of seven sub-periods, it is proved that my proposed strategy has better performance than the other two strategies do for gaining the profit.
2. After the examination of one complete period, it is proved that the most combinations of my proposed strategy have the most outstanding performance. Therefore, my proposed strategy can be recommended to the investors for their future application.
3. After the examination of this study period, it is proved that there are four combinations of the moving average strategy, seven combinations of filter rule strategy and four combinations of my strategy can be applied for future praxis.
This study further suggests that more alternative strategies be studied in the future, in hope to find out more potential strategies for excess return based on different cycles.
Key words: moving average, filter rule, trading system design, programming
|
|---|
