In his book ‘The 30-Minute Stock Trader’, Laurens Bensdorp describes a suite of 3 strategies, working in unison, and trading individual stocks from the S&P 500 universe. We implemented these strategies as a showcase.
The first strategy is a trend-following strategy, selecting stocks by their momentum. We enter new positions only when finding stocks at least slightly oversold. As a result, we improve the odds of trading profitably. There is no exit condition, as we will continuously rotate new or better candidates in.
The implementation is very straightforward, providing an excellent template for your own experiments.
Mean-reversion long/ short
The next two strategies are mean-reversion strategies for the long and the short side. These strategies are a bit more complicated than others of this kind, as we are using many more indicators: We filter by long-term momentum, recent volatility, ADX, and RSI. The strategy uses multiple exits, including time-based, stop-loss, and profit targets.
As the long and short sides of the strategy are very similar, we have implemented them together using a common abstract base class. This code nicely demonstrates how to separate the common strategy logic from a trader’s specific settings.
After implementing the strategy, we ran a bunch of simulations and optimizations of this strategy, based on a universe of about 180 stocks taken from S&P-100 and Nasdaq-100. We use lots of indicators in this strategy, which is further augmented by the large trading universe. As a result, this strategy is a bit more taxing than most of our other showcase strategies. However, our testing concluded that a typical desktop machine is well capable of simulating this strategy – after making some performance tweaks to TuringTrader’s indicator mechanism.
We hope that this showcase serves you as a solid template, and helps you spark some new ideas. If you are just curious and like to see how well these strategies do, head over to TuringTrader.com, where we have daily updated charts.