If someone wants to learn aspects of professional system development, I can recommend Daniel Fernandez' website:

http://www.asirikuy.com/

It's not free, and some methods are debatable, but so far it's one of the few websites with a serious and empirical approach to detail aspects of automated trading.

Only problem is that most of the information on the website is not easily accessible, but cluttered among hundreds of videos and lengthy forum discussions. So it's not easy to find the descriptions of algorithms and methods - you need time to watch weeks of videos and take notices.

Yes, It seems to be a serious site about automated trading.
But I don't know someone who is a part of it.

I had the opportunity to test some of his systems with Zorro. Most are not profitable, but a few are, although in the low range of profitability.

So you/someone of your team are/were a member of them.
Why they give their members non-profitable systems?
Will they profitable on their platform?
Let me know what platform they use. Maybe MT4 or/and Tradestation.

Do you know other community sites for programming systems?
Which I know of:
http://www.tradingsystemforex.com/
http://fxcodebase.com/
http://forex-tsd.com
http://www.stevehopwoodforex.com

Why they give their members non-profitable systems is easily answered: Daniel thinks that the systems are profitable. Or I think wrongly that they are not profitable. He has different profitability criteria and does not use walk forward tests. It's possible that I'm just over-critical. However, some were live trading his systems and lost money so far. Daniel earns his money with subscriptions to his website, not with trading.

He uses MT4 with a self-written strategy tester and a heavy additional framework for fixing some of the MT4 problems and shortcomings. The website has many different test tools written in many different computer languages.

Most other Internet trader communities are not very helpful for automated trading - their members just lack the knowledge and tools.

Are those fixes exclusive for them or provide with its members?

Did you build something like that in your Zorro?

Why most of MT4 robots fail is because of missing that?

The framework is available to the Asirikuy members. It is not needed for Zorro, as Zorro's own functions are comparable or better than that framework.

MT4 is not suited for serious automated trading. It's lacking essential features for test, script, and script execution. The first problem is the worst, because without a usable tester you're forced to trade untested and thus most likely unprofitable systems. That's why people wonder why they're losing money with their MT4 robots although they got high profits in the MT4 "strategy tester".

For manual trading, I think that MT4 is still useful.

Thanks for the clarification.

Another serious site for automated trading.
http://www.automated-trading-system.com/

Another one.
http://beathespread.com/

His blog is interesting.
http://www.rtnlblog.net

http://www.myforexdot.org.uk/

Trading Strategies & Trading Systems

Interesting.

http://systemtradersuccess.com/
http://rubberducky-techanalysis.blogspot.com

Thanks acidburn,

Has anyone any experience of using fractal dimension or efficiency ratios as a filter for trend trading strategies. In theory, they could make a chop filter.

I spent some little time jerking around with fractal dimension and hurst exponent quite some time ago. I'm afraid I couldn't conclude anything useful out of it. Mostly my results agree with what others report, it's very interesting to follow the theory behind HE & FD, but when it comes to practical application suddenly it becomes very hard or impossible.

Here's another link for you, now that you've expressed interest in this area, lots of good stuff here: http://www.bearcave.com/misl/misl_tech/wavelets/hurst/

Free monthly magazine (registration required)

http://tradersonline-mag.com/

Much appreciated acidburn.

Any links to good articles on inserting an extra 4-5 hours reading time into the 24?

Here's a couple of interesting links...

1. http://fractalfinance.blogspot.com
The author has written a few indicators for mql4 using Fractal Dimensions and Hurst Differential... They look quite interesting.

2. http://etfhq.com/blog/2010/05/25/best-technical-indicators
This links to a comparison of various indicators, including Variable Moving Averages and Adaptive Moving Averages, that use Efficiency Ratio & Fractal Dimensions, and a few others. They also test a modified Fractal Adaptive Moving Average (FRAMA) ("The FRAMA identifies the Fractal Dimension of prices over a specific period and then uses the result to dynamically adapt the smoothing period of an exponential moving average.")

All of which is so far over my head that it makes the international space station look like a submarine.