Simple Technical Trading Rules and the Stochastic Properties of Stock Returns
Read the paperopens doi.org in a new tab
What they found
The first widely cited academic test to take technical analysis seriously. The authors tested 26 moving-average crossover and trading-range breakout rules on the Dow Jones Industrial Average from 1897 to 1986. Buy signals produced average returns of about 12% annualized versus roughly minus 7% for sell signals, and buy-signal periods were less volatile than sell-signal periods. Using bootstrap methods they showed these patterns could not be reproduced by random-walk, AR(1), GARCH, or other standard return models.
What you can use
- Simple trend rules on a broad index showed real predictive power over 90 years, especially in distinguishing 'sell' regimes.
- The forecasting power was as much about volatility (sell signals precede rough markets) as about returns.
- The rules were common ones traders actually used, not data-mined combinations.
Caveats
No transaction costs; later work (Sullivan-Timmermann-White) showed the results shrink under data-snooping adjustment and vanish in post-1986 data. Index-level, not individual stocks.
Tags: technical-analysis, moving-average, breakout, bootstrap
Summaries are our own reading of the paper, not the authors' words. Educational only, not advice. Discuss it in Book Club.