Technical analysis indicators are widely used by traders in financial and<br>commodity markets to predict future price levels and enhance trading<br>profitability. We have previously shown a number of popular indicator-based<br>trading rules to be loss-making when applied individually in a systematic<br>manner. However, technical traders typically use combinations of a broad<br>range of technical indicators. Moreover, successful traders tend to adapt to<br>market conditions by ‘dropping’ trading rules as soon as they become<br>loss-making or when more profitable rules are found. In this paper we try to<br>emulate such traders by developing a trading system consisting of rules based<br>on combinations of different indicators at different frequencies and lags. An<br>initial portfolio of such rules is selected by a genetic algorithm applied to a<br>number of indicators calculated on a set of US Dollar/British Pound spot<br>foreign exchange tick data from 1994 to 1997 aggregated to various intraday<br>frequencies. The genetic algorithm is subsequently used at regular intervals<br>on out-of-sample data to provide new rules and a feedback system is utilized<br>to rebalance the rule portfolio, thus creating two levels of adaptivity. Despite<br>the individual indicators being generally loss-making over the data period,<br>the best rule found by the developed system is found to be modestly, but<br>significantly, profitable in the presence of realistic transaction costs.
