Abstract

This study evaluates the predictive power and economic effectiveness of ten variable-length moving average (VMA) and six trading range break-out (TRB) rules applied to S&P/TSX Composite Index. It is found that VMA rules do not generate random buy and sell signals, which is consistent with trading rules having predictive power. By capturing negative risk premia, the rules reduce risk and yield significant excess returns over a simple buy-and-hold-strategy. The effect of trading costs on profitability is nontrivial; break-even two-way transaction costs suggest the rules are effective only in a low-cost trading environment. The performance of the rules is consistent; the best and the worst performing rules in-sample remain so out-of-sample.