In addition to the popular futures and forex markets, many quantitative traders like to test and trade their strategies on liquid and active individual stocks.
AAPL is one of the most active stocks in the world, both in terms of volume traded and news coverage. It is also been on an incredible bull run over the past ten years. Any stock that has been as popular and newsworthy as AAPL has is bound to have some strategies custom designed for it.
Paststat.com devotes an entire section of their website to different quantitative ideas that traders are encouraged to take and develop for potential strategies. One of their recent articles featured an idea for a short-term breakout system designed specifically for trading AAPL.
The basic concept of the strategy is that it takes a long position in AAPL any time the stock breaks out and closes above its upper Bollinger Band. The strategy then holds the stock for between 1 and 5 days before selling. The strategy is based on a daily chart with Bollinger Band settings of 20 period moving average and 2 standard deviations.
The article describes the chart it contains very simply:
the trading odds for the $AAPL longs over the next 1/2/3/4/5 trading days period when ever $AAPL close crosses above the upper bollinger band
The article backtests this strategy from the breakout in December 2009 through November 2013. Considering the simplicity of the strategy, the results are actually quite impressive.
Using a hold period of one day, the strategy produces 22 winners out of 28 trades. The average profit on a winning trade is 1.04% and the average loss on a losing trade is 0.57%.
When the hold period is increased to five days, the strategy increased to 23 wins. The average profit on those winning trades was 2.69% and the average loss on a losing trade was 1.82%.
The article goes on to post the results of each of the trades that were logged for the five-day holding period approach over the course of the backtesting period. This really emphasizes the high number of winning trades and the fact that the winning trades are bigger than the losing trades.
While this is certainly a small sample size, the initial backtesting results indicate that it may be worth investing some further time to develop this into an actual strategy. It is possible that the returns could be improved on by implementing a trend filter or some other confirmation indicator.
It would be very interesting to see further testing of trading this strategy on AAPL. It would also be interesting to test it on other individual stocks, and to attempt to further refine it.