Hmm, I agree on some points and strongly disagree on others.
a) the definition of curve fitting was exactly my point. If you show that a statistical test is robust (and now we can discuss what robust means, but I mentioned how I understand it in my previous post) then I have confidence in the validity of the test.
b) You are right, everyone can show a fantastic back tested result. Question is, does it hold up to out of sample tests, and with out of sample I mean different time periods, not different assets. I strongly disagree with your blanket statement
"No system can be robust unless every market is traded in the EXACT SAME WAY."
Also your definitions of "trend system" are taken out of thin air. If I tested a trend system on currencies and it works well going forward, and that for a long period of time, while it did not back tested well nor performed well going forward on, lets say, crude contracts then I am very happy to apply my system to currencies and extract alpha.
I think you may still be confused about curve fitting. Curve fitting applies to the way you test, not on which data you test.
PS.: I never said my asset selection's purpose is to select those assets which I believe will trend. I said that my selection process's purpose is to only trade those assets that trend SIGNIFICANTLY WHEN they trend.
Of course can it happen that an asset that exhibited very poor trending properties will change its dynamics. Thats why I employ a rotational model which includes or excludes assets considered for position taking.
Quote from Ash1972:
Well, the term "curve fitting" is actually short for invalid (i.e. statistically non robust) optimisation. Obviously some back fitting will have been done, otherwise you will have no system at all.
Assigning properties to individual markets is statistically a type of prediction and a major mistake. Frankly, any amateur can create truly fantastic back tested results by taking 20 or so markets and fitting a couple of parameters on a *per market* basis. No system can be robust unless every market is traded in the EXACT SAME WAY.
What makes multiple markets useful is you can boost the risk reward ratio by using the fact that on the whole some markets will be trending whilst others aren't. This is meaningless if you've decided in advance which ones are going to trend. A trend following system by definition makes money by attempting to follow a trend WHEREVER it may occur.
WHAT IF your nicely mean reverting (for the last 20 years) market suddenly, totally out of the blue, became a trending market for the next 5 or 10 years? Your backfitted system will fail to capitalise on this. You will be in quite a bit of trouble too as no doubt you will have based your leverage on the smooth, attractive equity curve you have back fitted.
It's amazing how few people understand this.
