Technical Trading Rules and Market Efficiency: Evidence from the Australian Stock Exchange 1980-2002

This paper examines the validity of two classes of simple technical trading rules – moving averages and trading range breaks – in the Australian Stock Exchange. We conduct our empirical analysis in two stages. In the first stage, standard t-tests are used to compare returns generate by the technical trading rules against those generated by the buy-and-hold equivalent. In the second stage, we employ bootstrap methods to generate empirical distributions of trading returns simulated under four models of stock prices – the random walk with drift, AR(1), GARCH(1,1) and EGARCH(1,1) models. Using daily data from 1 January 1980 to 31 December 2002, we find that technical rules possess some predictive power over the full sample period. However, tests on four non-overlapping sub-samples reveal that the technical rules generate returns in excess of the buy-and-hold equivalent only in the pre-1991 sample period, but then begin to generate negative returns in the sample period post 1991. Results from the second stage tests also indicate that it is possible to reverse the standard test outcome of predictability by addressing non-normality, time-dependence and conditional heteroskedasticity in the data. Overall, our results finds the ASX informationally efficient and that the period post-1991 marks a significant rise in the stock market’s efficiency.