Technical analysis with artificial intelligence

Tyren · Sep 26, 2006

Some more info :
http://www.ensignsoftware.com/tips/tradingtips25.htm#Neural

fra10trader · Sep 27, 2006

Quote from mspkash1:

I've been using wealth lab to see if there are any TA based strategies out there that actually works...Haven't found any luck yet..I look for strategies that can perform better than S&P %00 or just simple buy and hold. The programming is based on PASCAL language. If you know little programming it'll be easy to pick up.

There's a great read by Brett Steenberger talking about testing TA based strategies based on 3 steps:

1) Identify high probability setups;

2) Track volume flows at the entry signal and immediately afterward to ensure that the setup is setting up properly;

3) Monitor volume flows during the trade to ensure that trade idea is following through.

the how and when are spelled out in the article: http://www.tradingmarkets.com/.site...How-to-put-information-behind-your-trades.cfm

fra10trader · Sep 27, 2006

Quote from ProfLogic:

I absolutely agree . . . based on what you have tested. Trust me when I say that there is a lot more out that then "run-of-the-mill" TA. Trust yourself and discover it.

Once you've exhausted your individual efforts at finding what works, it may have to see what others do to arm themselves in terms of tools and decision support. I dont think anyone has the silver bullet. Its more a method, as someone else pointed out, of repeating an approach that has a high degree of certainty in the outcome. Not every outcome will be a win - but time and the consistent application of the approach work in your favor.

take plenty of water for the journey ...

oh an example of some 'others': http://www.tradermike.net

marketsurfer · Sep 27, 2006

Quote from ProfLogic:

I absolutely agree . . . based on what you have tested. Trust me when I say that there is a lot more out that then "run-of-the-mill" TA. Trust yourself and discover it.

please tell us, professor. what are the latest TA tools, and I am not talking about renamed oldies.

surf

ProfLogic · Sep 27, 2006

Quote from marketsurfer:

please tell us, professor. what are the latest TA tools, and I am not talking about renamed oldies.

surf

Answered previously . . . posted and explained.
Where you are concerned . . . any explanation, no matter how detailed . . . is futile.

marketsurfer · Sep 27, 2006

Quote from ProfLogic:

Answered previously . . . posted and explained.
Where you are concerned . . . any explanation, no matter how detailed . . . is futile.

Lets discuss your supposed new ideas minus the hype and marketing BS. I am truly interested.

Thanks,

Surf

onlytronalso · Sep 27, 2006

Aren't the large firms already using A I ?
Program trading for the new millenium!

BlowFish · Sep 28, 2006

Do neural nets with real time genetic optimisation of inputs/outputs count as AI? Still underwelming results seems to me.

Cheers,
Nick

fra10trader · Sep 28, 2006

Quote from BlowFish:

Do neural nets with real time genetic optimisation of inputs/outputs count as AI? Still underwelming results seems to me.

Cheers,
Nick

uGH,

neural nets, ai - i am trying the odddsmaker from trade ideas and its seems the approach here is keep it simple, s-----. A quick and dirty but informative way of telling you whats working now. any one trying the beta from their blog?

onlytronalso · Sep 28, 2006

Quote from BlowFish:

Do neural nets with real time genetic optimisation of inputs/outputs count as AI? Still underwelming results seems to me.

Cheers,
Nick

Yes, and they have quantitative analysis now too.

Kind of scarey, considering what works for a while does not work forever UNLESS it's back tested a 100 years!

Most of these derivatives are relatively new. No way they can backtest them suffficiently. Makes for an inefficient market at times.

If they ever analyze the DOW for the past 100 years they might find something. This guy did!

http://caldaroew.spaces.live.com/