Thursday, June 17, 2010

Watukushay FE - An Intra Instrument Experiment - Part No.2

On yesterday's post we talked about Watukushay FE and the first essays to test the intra-instrument compatibility of this trading system. We arrived at some good results which showed that Watukushay FE did achieve long term profitability on all of these instruments while the yearly compounded profit to maximum draw down ratio was always much lower than for the EUR/USD. Since I left yesterday's post with some questions regarding the optimization and modification of Watukushay FE to improve results on these currency pairs today I will dedicate this post to the answering of these questions. I will talk to you about the optimizations I have done on the Watukushay FE system on different currency pairs, what their results have been and what they tell us about the inefficiency exploited by Watukushay FE and its potential to be used on several different instruments, effectively diversifying the usage of this trading strategy.

Many of you may be thinking that the results shared yesterday point to the fact that just a few coarse optimizations would lead us to a gigantic improvement in performance which would leave us with a multi-currency Watukushay FE system, something which is very reasonable to assume given the fact that systems usually improve significantly with only moderate optimization. However what reality shows us is somewhat different. The Watukushay FE trading system fails to improve significantly on almost all trading instruments despite coarse or even finer optimization which attempt to fine-tune the system's variables to each currency pair.

For example, the GBP/USD and USD/CHF achieve compunded yearly profit to maximum draw down ratios of about 1:5 and the usage of even fine grid optimizations is not able to bring this down to a 1:3 level. When we examine the trading characteristics of this system on these pairs we see that the overall market inefficiency exploited by the EA disappears for significant periods of time in which the instruments consolidate or experience wide ranges that are not at all favorable for Watukushay FE. It becomes obvious that the system is simply not able to cut loses short on these currency pairs with its present logic. However a few ideas have come to my mind to "adapt" Watukushay FE to the trading mechanics of these pairs and improve its trading characteristics, something which I will probably leave for a future post.

Perhaps the most important achievement of Watukushay FE comes when you analyze its results on the AUD/USD and USD/CAD instruments. With only very coarse optimizations on these currency pairs we are able to obtain marked improvements that show us that -for example- in the case of the AUD/USD only a modification of the RSI period grants a significant improvement of the compounded yearly profit to maximum draw down ratio from 1:5 to almost 2:1 in a ten year run. This shows that the modification of only one variable is enough to fit a system to another currency pair when the overall inefficiency has a similar presence than in the initial instrument it was designed for. Below you can see the backtesting results for the AUD/USD before an after an RSI period change, leaving all other variables -including SL and TP - intact.
The USD/CAD also improves its results incredibly only after small RSI period changes but results are improved much more significantly after the TP and SL are optimized in a coarse manner changing the risk to reward ratio of the SL and TP to almost 1:2, this in turn changes the compounded average yearly return to maximum draw down ratio to a little bit more than 1:1 from a previous value - before optimization- of nearly 1:4. The images below show you how the overall results for this currency pair also change as a function of this very limited optimization which is most likely not going to introduce any curve fitting into the system due to its coarse nature and the overall adaptive character of the trading system.
After doing this analysis of Watukushay FE it is also worth noting that the optimization of this EA are heavily populated by profitable results showing you that future deviations from optimal values are not going to drastically change the risk and profit characteristics of this trading system. We have also learned that when the inefficiency is present to a good extent only minor optimizations are needed to greatly increase performance while very fine and extensive runs -when the inefficiency is absent a significant amount of time- bring us only marginal improvement to pre-optimization results.

From the above results it seems that Watukushay FE will be exploring Canada and Australia soon enough :o) Probably during the next few weeks I will release EA modifications and set files on the experts website so that people can start doing their own evaluations and tests on these currency pairs. A couple of live accounts to test these new pairs will also be added. Right now I am also exploring logic modifications to trade on the GBP/USD and USD/CHF, results I will be sharing with you once they are ready. It seems that Watukushay FE likes more instruments than what I originally thought :o)

If you would like to learn more about Watukushay FE and how you too can build your own long term profitable system based on sound trading tactics please consider buying my ebook on automated trading or joining Asirikuy to receive all ebook purchase benefits, weekly updates, check the live accounts I am running with several expert advisors and get in the road towards long term success in the forex market using automated trading systems. I hope you enjoyed the article !

No comments:

Daniel Fernandez's Expert Author Email Alerts
Sign up to receive email alerts of Daniel Fernandez’s latest articles from!

Email Address: