Forex patterns

Quote from ronblack:

FIbs are good but you never know whether is going to be 50% or 61.8% for example, and that can cause a lot of pain if you are wrong.

Ron

Sure, nothing is right all the time which is why stops were invented :p

I was using Fibs as an example to show that thinking 'nothing that is obvious to you and to many others can make money' isn't exactly true.
 
Quote from cashmoney69:

I use fibs too, and its amazing how accurate they can be.

I'm amazed every day by how they work, if it was this easy everyone would be millionaires wouldn't they?

This is what gets me about (forex) traders, a lot of them think they need some secret strategy or that the more they pay for an indicator or system the better it will perform. Everything I use is free, didn't cost a dime.
 
Quote from ronblack:

Anyone using this program to trade forex?

http://www.tradingpatterns.com/Software/APS/forex/forex.html

The program -- apparently a data miner -- can find good numbers of patterns in daily and intraday forex data but I need to figure out a way of selecting those with the best potential to stay profitable.

Does anyone know of a suitable statistical test I can use?

Thanks, Ron


I keep it simple and use just the profit factor to screen systems. Some people may object to this but I have determined PF is one of the best filters.

Bill
 
Quote from acronym:

My god, this is the most shillworthy crap i have ever read in my entire life.

You gotta look harder, there's better shillworthy crap than this around :D
 
Quote from intradaybill:

There are other costs and restrictions:

"Platforms are priced at $60,000. A maintenance fee of $20,000 per year after the first 2 years is required for continued operation. The TSL license requires the signed acceptance of restrictions regarding external use of TSL generated systems or signals. "

http://www.tradingsystemlab.com/howtoorder.aspx

Sounds like a law firm with a software development department:)

Bill

That's way too expensive for what appears to be a program based on an inexpensive genetic programming engine.

Keep in mind that genetic programming is in principle a fairly simple idea. I believe it can only lead to good results when you start with a set of well-defined trading strategies. These "seeds" are not required to be profitable. What the genetic programming algorithm will do is try to find an "evolution" of those strategies that will produce good performance results.

I think such process is inherently very limited. It is also not very far from fitting. Automatic generation of trading systems is in its infancy. IMO data mining techniques like those used by Michael Harris can be more effective. He seems to be down to earth and AFAIK he is the first who developed a commercial product doing it:

http://www.tradingpatterns.com/About_Us/articles/synthesis.pdf

Alex
 
Here is a pattern of success.

Out of over 5,000 systems on c2, we have the top 5.

Out of those 5, the best tracked forex system continues to be RT Forex North.

An elite pattern and proof that some or at least one edge based system can actually trade forex profitably over a long time period.
 

Attachments

Quote from oraclewizard77:

Here is a pattern of success.

Out of over 5,000 systems on c2, we have the top 5.


Good for you but selection bais dictates that this is purely random.

People win the lottery after each draw very often but each one person has very low probabilities to win.

To say that someone won the lottery means nothing: it's just selection bias.

Good for them though.

Bill
 
Back
Top