Using a Differential Evolutionary Algorithm to Test the Efficient Market Hypothesis

Abstract

The proposition that a relatively new technology such as a differential evolutionary algorithm (DEA) can violate the weak form of the efficient market hypothesis is tested using daily data from the Australian share market from 2000 until 2008. An option trading strategy based on forecasts from a DEA is shown to perform better than a buy and hold strategy over parts of the sample space and, on average, over all of it. Speculators may make supernormal profits using new methodologies however such profits are unlikely to be sustained.