did i apply curve fitting to my system

[Mike805]

Quote from bwolinsky:

This all depends.

I agree, Mike, anyone with Gann in the name has no clue what he's talking about.

If your close to your original hypothesis, tweaking for an extra 10% APR or more is fine.

If you go from -50% NP to plus 1000%, there's a little too much fitting. If you're going from 70% APR to an extra 5% to 75-80% APR, there's nothing wrong with that.

It will depend on how much of a boost this is to your original results.

On your "if-touched" problem, only consider your order filled if the trade limit order "trades through" the price, or do what Mike suggested and add a tick of slippage plus commission to both sides.

You've pointed out the key to proper development - validation of the hypothesis.

If one does not know what their hypothesis is, or, they are searching for one via data mining, they are likely performing a destructive form of cruve fitting rather than actual constructive optimization.

Let me be very clear about this, curve fitting/optimization can be very useful if done properly. Also, changing one's sample space for OOS testing is paramount post optmization, i.e. walk-forward testing.

Truth be told I curve fit *everything* and every system I've gone live with has performed nearly identical to the model. The driving factor behind a system's success is the fundamental concept, not the fit params, unless of course those params void the hypothesis in some way.

 

[TSGannGalt]

Quote from Mike805:

Stop with the math teacher crap, its old. In line with your "my dick is bigger than yours" type comeback, I did my master's thesis on vola. modeling using something called an ASAM under the leading expert in AI and neural networks. What exactly are your credentials?

I'm a paper trader.

Because I'm Asian, I have a small penis.

 

[Mike805]

Quote from TSGannGalt:

I'm a paper trader.

Because I'm Asian, I have a small penis.

haha...

Somehow I don't believe you...

 

[TSGannGalt]

Quote from bwolinsky:

This all depends.

I agree, Mike, anyone with Gann in the name has no clue what he's talking about.

Anyone with "...insky" in their alias is a Russian Hacker trying to steal money from non-suspecting newbies.

 

[Mike805]

Quote from TSGannGalt:

Anyone with "...insky" in their alias is a Russian Hacker trying to steal money from non-suspecting newbies.

That, or he's Polish, which would be self-exaplanatory

 

[bwolinsky]

Quote from TSGannGalt:

Anyone with "...insky" in their alias is a Russian Hacker trying to steal money from non-suspecting newbies.

Both of you are right, Polish and Russian. You got me. Not a hacker though. If you did any amount of testing on Gann fans you'd find they don't work, and lead to the sort of curve fitting we're actually trying to avoid.

 

[TSGannGalt]

Quote from bwolinsky:

Both of you are right, Polish and Russian. You got me. Not a hacker though. If you did any amount of testing on Gann fans you'd find they don't work, and lead to the sort of curve fitting we're actually trying to avoid.

A Russian Polish Mob living in Chicago?

 

[trader3cnd]

Quote from Mike805:

You've pointed out the key to proper development - validation of the hypothesis.

If one does not know what their hypothesis is, or, they are searching for one via data mining, they are likely performing a destructive form of cruve fitting rather than actual constructive optimization.

Oh my, I wonder if you really know what you're talking about. I worked for a consulting firm that applied data mining to a wide variety of business applications, including loan data, automobile sales, etc.

I' ve never heard of "destructive curve fitting" or "constructive optimization" What is the problem with you people? Do you think before you start typing? Are you making up those terms to justify some type of cognitive misfunction related, most probably, to a type of autism?

Trading systems are evaluated not based on how they were developed but based on how they perform. I do not care if:

1. it is optimized to death
2. It is constructive optimization you used
3. It is destructiuve curve fitting that was apllied
4. whether you had sex while you were developing the system

I only care whether it works. You can find out after a few trades in most cases.

All else is talk for losers.

 

[Hugin]

OK, I'll take the bait...

Quote from trader3cnd:

I only care whether it works.

My guess is that we all do. People in this thread just don't want put in a lot of work into systems with close to zero probablilty of working.

Quote from trader3cnd:

You can find out after a few trades in most cases.

Really?
This is quite interesting... Care to tell us how you know that a system works after a few trades? By working I mean that it has a high probability of generating positive returns in the future.

/Hugin

 

[trader3cnd]

Quote from Hugin:

OK, I'll take the bait...

Don't get into the habit...

Quote from Hugin:

My guess is that we all do. People in this thread just don't want put in a lot of work into systems with close to zero probablilty of working.

Most people are lazy (1), don't have the required knowledge (2)and have no money to test systems (3). Miracles don't happen...

Quote from Hugin:

Really?
This is quite interesting... Care to tell us how you know that a system works after a few trades? By working I mean that it has a high probability of generating positive returns in the future.

What is quite interesting is that you are asking for free tutoring about key elements of success. You need to pay a lot of money to learn that, if you ever learn it... I can tell if a system has a high probability of generating positive returns in the mediuum term after 20 trades maximum. Usually after 10 trades.