Create and Test Forex Strategies
This shows you the differences between two versions of the page.
Both sides previous revisionPrevious revisionNext revision | Previous revisionNext revisionBoth sides next revision | ||
eas-guide:reactor [2019/04/06 14:33] – ?nolink in image tags vini | eas-guide:reactor [2019/05/02 13:05] – vini | ||
---|---|---|---|
Line 17: | Line 17: | ||
The final goal of the tool is to provide strategies that have passed all robustness tests. So we can start EA Studio for a night and we will have ready for trading strategies set at the morning. | The final goal of the tool is to provide strategies that have passed all robustness tests. So we can start EA Studio for a night and we will have ready for trading strategies set at the morning. | ||
- | ===== Optimization | + | ===== Optimize strategies |
+ | |||
+ | ==== Full Data Optimization | ||
{{ : | {{ : | ||
Line 27: | Line 29: | ||
You may find it useful to increase the Generator' | You may find it useful to increase the Generator' | ||
- | ==== Walk Forward ==== | + | ==== Walk Forward |
**Walk Forward analysis (WFA) or Walk Forward optimization (WFO)** is a sequential optimization applied to an investment strategy. The name of the analysis is called “walk forward” because we have a moving window that progressively traverses the whole period of the data history with a pre-established step. Algorithmic traders apply such analysis to decrease the over-optimized parameters used in the investment strategy as we don’t want only a great looking backtest result, we want also a system that don't fail in live real money account. | **Walk Forward analysis (WFA) or Walk Forward optimization (WFO)** is a sequential optimization applied to an investment strategy. The name of the analysis is called “walk forward” because we have a moving window that progressively traverses the whole period of the data history with a pre-established step. Algorithmic traders apply such analysis to decrease the over-optimized parameters used in the investment strategy as we don’t want only a great looking backtest result, we want also a system that don't fail in live real money account. | ||
Read more about it at [[eas-guide: | Read more about it at [[eas-guide: | ||
- | ===== Monte Carlo Validation | + | In the Reactor context you can apply this optimization to the strategy, if the WFO results (Out of Sample net) perform better than the original strategy (full data backtest) the last parameters used by the optimization will be accepted to the strategy. |
+ | |||
+ | ===== Perform robustness testing ===== | ||
+ | |||
+ | ==== Walk Forward validation ==== | ||
+ | You can use WFO as validation tool and it returns only strategies that pass the validated segments without changing the original parameters of the strategy. | ||
+ | |||
+ | ==== Monte Carlo Validation ==== | ||
{{ : | {{ : | ||
Line 41: | Line 49: | ||
You set the percent rate of the tests on profit. In the shown example, all simulations must finish on profit in order to pass the strategy to the next tool. | You set the percent rate of the tests on profit. In the shown example, all simulations must finish on profit in order to pass the strategy to the next tool. | ||
- | ===== Multi Market Validation | + | ==== Multi Market Validation ==== |
{{ : | {{ : | ||
Line 49: | Line 57: | ||
We have 6 markets on the example above and the validation threshold is 3. That means that if 3 or more tests finish on profit, the strategy will ascend to the Collection. | We have 6 markets on the example above and the validation threshold is 3. That means that if 3 or more tests finish on profit, the strategy will ascend to the Collection. | ||
+ | |||
+ | ==== Hint for Reactor / Validator ==== | ||
+ | It is best to use only one of the following tools: | ||
+ | * Full Data optimization | ||
+ | * Walk Forward optimization | ||
+ | * Walk Forward validation | ||
~~DISQUS~~ | ~~DISQUS~~ | ||