Testing the profitability of moving-average rules as a portfolio selection strategy

One of the main difficulties in evaluating the profits obtained using technical analysis is that performance of trading rules depends upon the judicious choice of rule parameters. In this paper, popular moving-average (or cross-over) rules are applied to a cross-section of Australian stocks and the signals from the rules are used to form portfolios. The performance of the trading rules across the full range of possible parameter values is evaluated by means of an aggregate test that does not depend on the parameters of the rules. The results indicate that for a wide range of parameters moving-average rules generate contrarian profits (profits from the moving-average rules are negative). In bootstrap simulations the return statistics are significant indicating that the moving-average rules pick up some form of systematic variation in returns that does not correlate with the standard risk factors.