DLPAL trading Strategy for VXX volatility ETF has robust performance with 103% CAGR and relatively low drawdown in out-of-sample validation.
The in-sample for training and test is from 01/30/2009 to 12/31/2015. The out-of-sample for validation is from 01/01/2016 to 06/01/2017.
Below is the DLPAL S workspace used for in-sample unsupervised feature identification and supervised train and test. Profit target and stop-loss are set at 5%
The results are shown below:
Each line in the results is a strategy that satisfies the performance parameters specified. Index and Index Date are used internally to classify strategies. Trade on is the entry point, in this case the Open of next bar. P is the success rate of the strategy, PF is the profit factor, Trades is the number of historical trades, CL is the maximum number of consecutive losers, Type is LONG for long strategies and SHORT for short strategies, Target is the profit target, Stop is the stop-loss and C indicates whether % or points for the exits, in this case it is %. Last Date and First Date are the last and first date in the data file.
DLPAL found 110 short strategies in the in-sample. Below is the equity performance in-sample and out-of-sample with DLPAL code generated for Amibroker platform. Backtest includes $0.01 per share commission. Equity was fully invested with multiple signals ignored and initial capital was $100K.
Out-of-sample win rate is 62%, CAGR is about 104% and maximum drawdown is less than 25%. Total return is 172%, 90% in 2016 and 43% year-to-date. This is significant performance as compared to a short and hold strategy with equity curve shown below:
CAGR is 53% for short and hold at -40.2% maximum drawdown. MAR in out-of-sample is 1.32 for short and hold versus 4.32 for the DLPAL strategy. This is an indication that the strategy may be significant.
Note that 5% was the only level tried for the exits and there were no other trials involved. If many trials are involved, as it is usually the case with relentless data-mining for the purpose of developing a seemingly good strategy, then with high probability performance is artifact of data snooping. For more details see this article.
Edit: Below are backtest results in Quantopian platform 1-minute data and with code generated by DLPAL. The total return and maximum drawdown figures are similar.
It will be interesting to see whether this strategy will hold well if volatility increases.
DLPAL S is only $575 for the first year and if the license is renewed for a second year at the same guaranteed price, then the license is converted to lifetime. Click here for more details.
If you have any questions or comments, happy to connect on Twitter: @priceactionlab
Disclaimer: No part of the analysis in this blog constitutes a trade recommendation. The past performance of any trading system or methodology is not necessarily indicative of future results. Read the full disclaimer here.