Optimizing the forex trading system parameters: AUD/USD - Optimizing the trade entry parameter

User Rating: / 2
PoorBest 
Written by Forex Automaton   
Thursday, 14 May 2009 14:12
Article Index
Optimizing the forex trading system parameters: AUD/USD
Optimizing the forecasting parameter
Optimizing the stop-loss parameter
Optimizing the trade entry parameter
Optimizing the trade exit parameter
Summary of progress
All Pages

Trade entry threshold

Dependence of the annualized return on the trade entry parameter. Quality cut progression. AUD/USD. Dependence of the Sharpe ratio on the trade entry parameter. Quality cut progression. AUD/USD.

Fig.3. Top: Profile histograms showing dependence of the logarithmic measure of mean annualized return on the trade entry threshold parameter. The unit of annual return is 100%. Bottom: same for the Sharpe ratio. The system of performance quality cuts is as discussed above.

The entry and exit threshold parameters have been explained elsewhere.

With this definition of the performance quality cut, the performance quality requirement, synonymous with low drawdown, is seen to favor high values of the entry threshold. Model-wise, such high values correspond to the behavior of a conservative trader who only responds to high stimuli. This is consistent with conclusions from the previous reports. The shapes of the dependencies themselves seem to vary considerably from report to report, making it questionable to use them in parameter selection.

Dependence of the annualized return on the trade entry parameter. Best other parameters. AUD/USD. Dependence of the Sharpe ratio on the trade entry parameter. Best other parameters. AUD/USD.

Fig.4. Top: Profile histograms showing dependence of the mean annualized return (an average of monthly statistics) on the trade entry threshold parameter. 1 is 100% per annum. Bottom: same for the Sharpe ratio. Both figures are by the best Fred and stop-loss selection established in the previous sections.

In Fig.4, I remove the quality cuts and replace them with a cut based on the previously selected values of the forecasting parameter and stop-loss. The hope is that this latter cut confers extra quality to the procedure, moving the "context" in which the optimum for the entry parameter is sought closer to the production environment.

From Fig.4, the reader can get a feeling for the range of returns we are looking at. The relatively low Sharpe ratio of 0.3-0.4, typical for these reports, should not be discouraging since in the production regime, results for multiple different currency pairs will be combined and part of the fluctuation will cancel out, much the same way stock portfolios perform better than individual stocks. The exact degree of such cancelation is a serious question requiring special study.

Here, both return and Sharpe show quite peculiar response to the trade entry threshold and I would not base decisions regarding the trade entry threshold on the shapes in Fig.4. I am choosing the entry threshold range 0.004-0.006.



Last Updated ( Monday, 04 January 2010 12:35 )