My expert advisor auto-trading fund is currently in drawdown by the end of day 4. This forced me to make a detailed analysis of all the closing trades since day 1. I have realised that the GBP/USD is actually underperforming and it is mostly responsible for all of the incurred trading losses.
Optimisation Results
To troubleshoot, I performed some backtesing for the period (1 Oct to present). The results have suggested that I might have overlooked 1 or 2 important system parameters for the GBP/USD market, i.e. the system win-rate and the stop-loss margin.
The following results are based on 1 full lot trading at any one time and with 5% risk per trade:
# Original EA with 30 pips stop loss
(65 Trades / 49.23% win-rate / Overall growth 125.8%)
# Optimised EA with 40 pips stop loss
(58 Trades / 58.62% win-rate / Overall growth 128.9% )
# Optimised EA with 50 pips stop loss
(55 Trades / 63.6% win-rate / Overall growth 91% )
Obviously, the recent 3-4 trading days have been hard for the GBP/USD. Based on these results alone, I made my first parameter change in the EA. The reasons are simple: to achieve a higher system win-rate and a smoother equity curve.
Future Reference
For the purpose of comparison only, my first EA will actually put me in the Top 15 of the 2007 Automated Trading Championship at this point of writing. Note that the participants are allowed to open up to 3 trades simultaneously. I will need to incorporate that in the future.
Last but not least, here is the equity curve of current leader in the competition.
(For the same period, EUR-USD instead of GBP-USD / 270 Trades / 68.5% win-rate / Overall growth 765% )
No comments:
Post a Comment