How do you gain, and measure, your trading system's statistical edge?

http://www.forexfactory.com/showthread.php?t=22287

For anyone with an edge and who is also looking for the standard of risk adjusted return calculations.

Thanks, I will use that if/when I try an incorporate this edge into a complete trading system.

That formula is too complicated and doesn't make sense before you start trading. As long as I can backtest 10 years of data and I know the maximum drawdown and the potential win then I'll look if the 2 * maxDD and win / 2 are still ok for my taste. If I'd start to trade it I'd consider 2 * maxDD to be the maximum accepted loss.

Ds, that's bs. I don't say it lightly because you're one of my favourite posters on here. But a system is the combination of an inefficiency- an edge- plus trading rules. If you jump straight into dd and rar before you've even proven if your edge is statistically valid then you'll muddy your stats right off the bat.
Some edges can be proven but don't easily fall into a system. Some systems are profitable for a certain amount of time but don't have an edge- which means failure is round the corner.

This guy is asking if his edge is valid, not if his system is worth trading. Very different Imo.

The OP has asked for a hand in determining the statistical edge of systems or more setups as i see.

So, in this case the drawdown IS NOT a concern, can we see why?

As it is, a traders personal objectives are what will determine how much or to what degree drawdown concerns him. Like for example Acumen (whose work i respect btw) is looking for his drawdown or earn figures on a monthly basis and maybe you are on a yearly.

If you are going to reinvest profits and compound in order to attain max terminal wealth at the end of a set time frame, then drawdown is essentially meaningless. Think about it, how many heads in a row (or tails) in a 1000 tosses, how many such streaks? At the end if ur +ev, ur +ev. Its the biggest loss which tells us exactly what we can afford to get away betting (optimally, for some) without getting our head handed to us. Basically drawdown as a stat is inherently not what traders should concentrate on, 10 losses in a row dosent mean ur system is done (If you have worth any merit to begin with that is).

I know that people will object that everyone is not out for max wealth (and logically so), and i agree, but that is a function of each persons utility (which varies a lot), so we cant objectively decide on any 'one' utility curve. Someone can be a gunslinger, the other a monk.
So the systems edge is independent of drawdown. Biggest loss gives the story (but no stops so ..).

The above is just my opinion.

Could you maybe share the indicator or what ever you use to calculate the edge automatically? I don't think you have done it by hand or am i wrong?

This statement is true but misses the point of why you should do it. If the system was profitable using 80 days but only half as profittable at 75 and 85 then you have by chance curve fitted. The same goes for the 20 day holding period rule.

The sensitivity of the results to the input parameters is an important consideration in deciding if the there really is a robust system here. A robust system will work reasonably well over a range of values. If it only worked for a few values I would consider the results to be unrepresentative.

An earlier post also made the point about this not working for other pairs. I would also test it on a similar pair like GBP/USD becuase if it same parameters did not work reasonably for GBP it would be another suggestion that the 80 day value is a fluke.

We tend to use round numbers for simplicity, I'd be surprised if anyone would start his test at 83 days.
As the best number of days is a nice case of curve fitting, a better idea is to take a fourth of profit aftter 60, 80, 100 and 120 days.

I don't think the statistical edge is a direct constant. I think it changes given market conditions. I think you would have a hard time finding a system that maintains a constant edge over any serious amount of time.

In some cases your system edge may be high where as other times it immediately goes to DD.

Because markets change, so will any statistical edge with it.

Given that tho' I like the conversation. I may have gleaned at least one new idea from it. Also any system should be made so that you personally can live with it. If you shoot for a modest 2 or 3% a week, but are content with that then you "may" shave some DD off by not being too greedy.

When you don't worry about the "extra" 2 or 3% more that you could have gotten I think that is when you become a more confident trader.

Ya know I been thinking about this over the past day or so and the thing I get hung up on is that darn fixed point in time in the future.

I think the fixed point in time shouldn't be a fixed point at all. I think it should adjust as the need be. But I am not sure how to go about it just yet.

As a primary example if you have a fixed point in time and the time over laps a news event the results may or may not be accurate as opposed to other less moving time periods.

Additionally looking back at other analysis I have done in the past will show activity as to when gains were hit / closed out vs losses taken etc. Some days were more or less blank with no activity at all... in other words it varied.

So I think to properly forecast the edge you would need some sort of argument that that would shift the fixed period of time as needed, thus trying to maintain the edge at all times.