Title :
Trading the stock markets using genetic fuzzy modeling
Author :
Ettes, Drs Dennis
Author_Institution :
Dept. of Electr. Eng., Delft Univ. of Technol., Netherlands
Abstract :
In this paper the reliability of trading systems that use prediction models as a rating model of stocks is analyzed and an improvement is proposed. It was found that the RISE criterion is unreliable for selecting a profitable model. To increase the reliability of the trading systems an approach is proposed that optimizes the rating models with a profit goal. Two different profit goals are tested. The first is a direct profit goal and the second goal includes sensitivity analysis in the goal function. In both cases, the rating models are optimized using a GA. The simulations show that the results are better and more reliable
Keywords :
digital simulation; financial data processing; fuzzy logic; genetic algorithms; sensitivity analysis; stock markets; RISE criterion; direct profit goal; genetic algorithm; genetic fuzzy modeling; optimization; prediction models; sensitivity analysis; simulations; stock market trading; stock rating model; trading system reliability; Data preprocessing; Electronic mail; Fuzzy sets; Genetics; Laboratories; Portfolios; Predictive models; Sensitivity analysis; Stock markets; Testing;
Conference_Titel :
Computational Intelligence for Financial Engineering, 2000. (CIFEr) Proceedings of the IEEE/IAFE/INFORMS 2000 Conference on
Conference_Location :
New York, NY
Print_ISBN :
0-7803-6429-5
DOI :
10.1109/CIFER.2000.844591
