Finally devoting some time exploring Chaos Algorithms on a range of minute charts and some striking systems have become apparent One advantage of minute charts from a business "sales" perspective is that one can demonstrate longish performance histories tends to strenghten subscriber resolve in holding and trading the system through any drawdowns, because they have a longish backtest history to analyze and gain confidence from
In the past 1-18 minute charts while profitable were nowhere close to the tick charts performance and much of the research took place on tick charts that yielded striking performance in comparison. Thus ESThunder, ESLightning, CLFlow, CLFire, in fact all of them are on tick charts
Recently started exploring a larger range of minute charts, especially the higher time-frame, 29 minute and up and was a revelation with numerous profitable systems it became a matter of cherry picking the best with literally thousands of permutations and combinations of profitablity.
These are NOT brute forced hyper optimized curve fitted systems but are rather "naturally" profitable upon application to the chart. Then it becomes a matter of finding the best profit/stop loss value combinations Back testing in such a situation is as simple as extending the chart time-frame backwards by a few more years and seeing the results
Out of this process arrives astonishing CLFlame trades on the 123 minute oil as close to a linear hypothetical performance curve as one can imagine charted on all available significant oil volume past history (Beyond 8 years is too thinly traded with many gaps in the chart)
In the past 1-18 minute charts while profitable were nowhere close to the tick charts performance and much of the research took place on tick charts that yielded striking performance in comparison. Thus ESThunder, ESLightning, CLFlow, CLFire, in fact all of them are on tick charts
Recently started exploring a larger range of minute charts, especially the higher time-frame, 29 minute and up and was a revelation with numerous profitable systems it became a matter of cherry picking the best with literally thousands of permutations and combinations of profitablity.
These are NOT brute forced hyper optimized curve fitted systems but are rather "naturally" profitable upon application to the chart. Then it becomes a matter of finding the best profit/stop loss value combinations Back testing in such a situation is as simple as extending the chart time-frame backwards by a few more years and seeing the results
Out of this process arrives astonishing CLFlame trades on the 123 minute oil as close to a linear hypothetical performance curve as one can imagine charted on all available significant oil volume past history (Beyond 8 years is too thinly traded with many gaps in the chart)
No comments:
Post a Comment