Thinking of a system. | Forex Factory

Options

Permalinks
Print Thread

Similar Threads

Fibnacci trading..Are U thinking what i'm thinking? 76 replies

Am I thinking about this right? 2 replies

Start Thinking In Terms of Risk-Reward, by Dr Van K. Tharp 0 replies

Put your thinking cap on 2 replies

Thinking of a system.

  • #1
  • First Post: Oct 10, 2006 12:33pm Oct 10, 2006 12:33pm
I think this is a wierd question. I was talking to my buddy about forex because he was wondering what I'm always looking at. We got into it pretty deep and he started asking about the different kind of systems and whatnot. He asked me a question and I didn't know how to respond....

Basically he wanted to know if you could take moving averages and basically plot every period to see which crosses produce the best results. See which ma's crossover's best predict a short or long position. It seemed like a logical idea to me. What does everyone think?
  • #2
  • Oct 10, 2006 12:38pm Oct 10, 2006 12:38pm
You can optimize MA parameters. However, this gives you the best performance in the past. It doesn't tell if it is the best choice for the future. I use just 10/25/50.

Perhaps 11/30/52 is better, i dont know.

I just would not take too short or too long MA's.
  • #3
  • Oct 10, 2006 12:45pm Oct 10, 2006 12:45pm
You can, and most of us programmers have.

If you read Ryan Jone's "The Trading Game" he does just that... But just know that MA's are optimized against OLD data; and it changes every year, and depends on the market. In fact, the optimized MA for this month probably aren't the same for next month; so it's chasing a carrot in front of you, that you'll never get.

That said, a "system" involves quite a lot...

You seem to be talking about "an entry and possibly an exit".. which does not a system make..

Position size, dealing with stress, strict guidelines of following the system, frequency, payrate, winrate, timeframe blah blah blah I can go on and on..

The market is like a woman; difficult if nigh impossible to understand; you can only perhaps understand the environmental impact of their actions, you can never predict their actions. You can only make sure you shore up those stops, and try not to let it get to you, when it behaves irrationally for no good reason. If the market behaves erratically, just chant to yourself "Women and shoes; women and shoes.." I find it helps me accept why the market doesn't make sense..
google:
  • #4
  • Oct 10, 2006 12:48pm Oct 10, 2006 12:48pm
Geeze Twoblink, You can't be winning many female friends with that statement... LOL

scott


Quoting twoblink
You can, and most of us programmers have.

If you read Ryan Jone's "The Trading Game" he does just that... But just know that MA's are optimized against OLD data; and it changes every year, and depends on the market. In fact, the optimized MA for this month probably aren't the same for next month; so it's chasing a carrot in front of you, that you'll never get.

That said, a "system" involves quite a lot...

You seem to be talking about "an entry and possibly an exit".. which does not a system make..

Position size, dealing with stress, strict guidelines of following the system, frequency, payrate, winrate, timeframe blah blah blah I can go on and on..

The market is like a woman; difficult if nigh impossible to understand; you can only perhaps understand the environmental impact of their actions, you can never predict their actions. You can only make sure you shore up those stops, and try not to let it get to you, when it behaves irrationally for no good reason. If the market behaves erratically, just chant to yourself "Women and shoes; women and shoes.." I find it helps me accept why the market doesn't make sense..
  • #5
  • Oct 10, 2006 3:16pm Oct 10, 2006 3:16pm
Quoting -25-
You can optimize MA parameters. However, this gives you the best performance in the past. It doesn't tell if it is the best choice for the future. I use just 10/25/50.

Perhaps 11/30/52 is better, i dont know.

I just would not take too short or too long MA's.
I'm curious to know why you chose those nunbers since you are anti-optimization.
  • #6
  • Oct 10, 2006 3:18pm Oct 10, 2006 3:18pm
Quoting philmcgrew
I'm curious to know why you chose those nunbers since you are anti-optimization.
They are not optimized in any way, i just chose them. I find 10,25,50 nice numbers.
  • #7
  • Oct 10, 2006 3:22pm Oct 10, 2006 3:22pm
Quoting -25-
They are not optimized in any way, i just chose them. I find 10,25,50 nice numbers.
You just pulled them out of your hind quarters and they magiaclly worked? Most interesting.
  • #8
  • Oct 10, 2006 3:47pm Oct 10, 2006 3:47pm
I think that optimization could only serve its purpose within a specific range just to get the values that would make trade work. But historical data used for optimization is certainly not going to be blamed when your trade results suck in the future. Optimization could only get you thus far.
  • #9
  • Last Post: Oct 10, 2006 5:05pm Oct 10, 2006 5:05pm
take a MA cross ea and run it in optimization mode in mt4. you can set it to run the fast ma from 1 to 100 and the slow from 1 to 100. it will probably take a few days tho
0 traders viewing now