THIS IS JUST FOR MB (email them yourself)
trading solutions in my opinion is very good but unless you are experianced in function modeling or abstract thinking in building your formula models it’s just like any other neural net.
Paying $2500 to just take a normal indicator and tinker with it, is quite foolish,and you can do that with any free software. With one ecception with trading solutions you can build an indicator of an indicator and thereby create your own indicators then test them. BUT STILL TO EXPENSIVE FOR JUST THAT
BUT, If you write your own formulas in abstract codeing that looks like your doing physics equasions then maybe.But either way you have about 1-2 years of studying to do before you know what is really possible, AND THEN you can learn how to tinker around with the neural net side and the way it calculates and learns, so add another 1-2 years.Then you can buy add ons that corrilate/combine with matlab and by the time your done learning, in about 5 years you can get back to us and tell us if it’s any good.
I GUESS YOU COULD LOOK AT IT AS SOMETHING TO KEEP YOU OUT OF TROUBLE AND AWAY FROM GETTING DRUNK AND GETTING ALL THOSE DUI’S YOU’VE BEEN GETTING LATELY,”RIGHT”SO SEE YOU IN 5 YEARS

P.S. MB don’t be afraid to go out there and take on responabilaty it will only make you a responsible person that wants to live in the real world.not someone that wants to aviod it NOW START THOSE EMAILS ROLLING!

http://www.altreva.com/

The Fundamental model is tricky to build as you’ve already found out. In my beta model, I downloaded Yahoo fundamental data using a quote donwload software and then using GA’s found out which parts of the fundamental data drove the year end targets. These year end targets were determined by “experts” so I take them with a grain of salt, I would probably use a discounted cash flow model to calculate the 1 year target myself but I didn’t have time.

Once I learned which pieces of fundamental data drove the end price, I then plugged them into Yahoo Stock Screener to find which ones match my criteria.

I found Noble Drilling (NE) that way and have been long it since May(?) sometime, its a good earner so far.

http://sourceforge.net/forum/forum.php?forum_id=390413
http://yale.sourceforge.net/wiki/index.php/Main_Page

Tom, do you actually use this model? It doesn’t seem very useful as it is. Who decides the 1 year target? Once you have your model, how do you decide your top picks? Where do you get your fundamental data feed? I’m also trying to use RapidMiner to create a trading model as well. But I’m stuck learning all about data mining & RapidMiner quirks.