In this post we review the performance of DLPAL LS generated features in 2018 applied to dollar neutral long/short strategy with daily daily data of Dow-30 stocks.
The performance is based on a simple rank using the features without any additional processing because that can be perceived as curve-fitting. Examples of how the features are generated by DLPAL LS can be found in this article. Briefly:
DLPAL LS uses primitive attributes of price action to extract features types in an unsupervised learning mode based on general feature clusters. Then, the program uses the extracted features in supervised learning mode to identify long and short candidates in a universe of securities. More…
DLPAL LS generates various types of features for individual but also for groups of securities but the two relevant for long/short equity are the directional bias Pdelta and its significance S. The strategy is as follows:
Time-frame: Daily (adjusted data)
Strategy type: Dollar neutral long/short equity
Universe: All Dow stocks from current composition
Backtest period: 01/02/2018 – 12/31/2018
History considered for feature calculations: 01/03/2000 – 12/31/2018
Maximum open positions: 6 long and 6 short
Position size per stock: Equity/12
Position entry: Open of next bar
Position exit: Open of next bar
Score for security ranking: Pdelta × S
Buy the top 6 and short the bottom 6 stocks based on score at the open of next day
Below are the results for the strategy and a comparison to buy and hold. We have also included results for a variation of the strategy that trades maximum 3 long and 3 short with Equity/6 position size. There is no use of margin and commission is set to zero.
|Strategy 6/6||Strategy 3/3||SPY TR|
Commission and slippage may vary from 1.5% up to 3% depending on specific conditions.
Below are the equity curve, underwear equity curve and monthly returns for the main strategy.
Despite the volatility the strategy performed well even during a downtrend for the market. Again, this strategy was based on simple score without any additional filters and machine learning to increase performance. In other words, the performance reflects the economic value of the DLPAL LS features before any boosting is applied.
Click here for more details and for information on requesting a free trial.
If you have any questions or comments, happy to connect on Twitter: @priceactionlab