Genetic programming: on the programming of computers by means of natural selection
Genetic programming: on the programming of computers by means of natural selection
Explorations in LCS Models of Stock Trading
IWLCS '01 Revised Papers from the 4th International Workshop on Advances in Learning Classifier Systems
Generating trading rules on the stock markets with genetic programming
Computers and Operations Research
Classifier fitness based on accuracy
Evolutionary Computation
KES'05 Proceedings of the 9th international conference on Knowledge-Based Intelligent Information and Engineering Systems - Volume Part I
Hi-index | 0.00 |
In the recent years the automatic generation of trading rules for stock and currency markets by means of Evolutionary Algorithms has become a popular game. Although, it is disputed whether or not such evolved trading rules are able to generate reliable profit on out-of-sample sets, especially if trading costs are considered. In this paper we focus on tickwise data and introduce a simple trading scheme based on Learning Classifier like action rules. These rules have only access to the most recent time series history and are thus only able to exploit the short term memory effects of tickwise data. Rather than searching for profitable trading rules on tickwise data, we first concentrate on evaluating the predictive properties of alternative indices, namely moving averages.