DLPAL S was used to design a strategy for Bitcoin futures using in-sample data. The strategy was then tested out-of-sample and found to have performed well.
DLPAL S setup
We used daily backward-adjusted data for Bitcoin futures from 10/18/2017 to 12/31/2020 as the in-sample. Below is the DLPAL S workspace used for the strategy search.
The Extended search cluster (default) was used and all parameters were set as shown on workspace above. The profit target and stop-loss were set to 5%. This workspace was used in a test to determine whether DLPAL S could identify strategies that perform well out-of-sample.
Results
DLPAL S identified a total of 18 strategies, 6 long and 12 short.
Equity curve
Below is the equity curve of the system of the 12 strategies for combined in-sample/out-of-sample. The out-of-sample covers the period 01/04/2021 to 06/29/2021. After the results were generated, DLPAL S configured the system using all identified strategies (to minimize selection bias) and generated code for Amibroker. Note that there are many possible ways to combine the strategies but in this case we used the simplest one based on the OR Boolean operator.
Out-of-sample return is 72%. Win rate is 53.5%. Number of trades in out-of-sample: 45 (20 long and 25 short).
Normally, the user may need to run DLPAL S every 6 to 12 months to rebalance the strategy. The above strategy is just an example and not recommended.
Notes
- Initial capital for in-sample and out-of-sample is $100K
- Margin for futures is $13,574
- Only one contract was traded per position.
- Out-of-sample performance may be due to luck.
- There is no guarantee any strategies will remain profitable in future.
More articles about using DLPAL S can be found here. You can request a free demo of DLPAL S from this link.
Charting and backtesting program: Amibroker. Data provider: Norgate Data
If you found this article interesting, you may follow this blog via RSS or Email, or in Twitter
