developing a automated trading system

[FibbGann]

Rimm

[its1111]

Fibbgann

Looks like it catches things very nicely. Good for you to develope something like this. I am sure it is a lot of hard work, when it comes out, please let us know if you are interested in selling it.
Mark

[FibbGann]

Here are some result for 120 days of backtesting the ES on a 3 minute. Using Strategy.MARKET NEXTBAR.

1 contract

Screenshot1:

[FibbGann]

Screenshot2:

[FibbGann]

Screenshot3:

[FibbGann]

Screenshot4:

[FibbGann]

Today's triggers

[FibbGann]

Its1111 I sent you a PM let me know.

[Slipperx]

This looks fantastic but I wonder if you have been able to backtest in over different market conditions than just the last three months. Having said that looking at the smaller timeframe charts I shouldn't think there is much of an issue.

One thing if autotrading is that you would probably want to be out at the end of the day so as not to be holding overnight orders (if trading the smaller time frames). I guess that should be easy enough to add in. I would love to see the code

I am working on a Pivot strategy for the Dow and eminis and have the same backtesting problem as above. In years of trading I have found whenever I have developed a system that looks great and implement it the market throws me a curveball and changes in nature - guess that's whats known as overoptimising something - which is why I mentioned the backtesting study sample issue above. Great work - well done.

[demarcog]

Congratulations on an incredible system

I've been working full time for the last 5 years tyring to develop something that makes money on the S&P in 1-5 minute data and have basically just about given up hope of ever coming up with something worth trading.

The results you posted were nothing short of amazing, I was beginning to doubt that it was possible but you apparently have succeeeded. I have systems that "tested" as well, but was using Strategy.Market and actually getting filled at the open, not "market" and in a real market all my "profits" disappeared so make sure you verify realtime that the fills given to the strategy analyzer are actually real.



Any insight, or information regarding you system or suggestions for development that you are willing to share with those of us less successful in the trenches would be greatly appreciated. What indicators have you found useful? Do you recommend stops and profit targets being coded into strategies.

Thanks alot and keep up the good work... [email protected].

[Richard Yu]

535 trades and $831 net profit assuming zero slippage and commission.... With realistic slippage and commission, the result will suffer..... However, the % of winners is really impressive. Maybe it can be more profitable with less trades if the timeframe is longer....

[demarcog]

Point is well taken, I can't believe I missed the no slippage and commission.

It's been so long since I'ver tested a trading system without including slippage and commission and in my testing on the E-mini's mean a minimum of $15.00 that I mistakenly assumed the results were net results.

[FibbGann]

Valid points everyone.

I did not intentionally leave out the commission, It was just a personal habit because I do it in my head.

After trading the S&P E-mini's for a long-time I have dropped them and moved onto the Russel Emini's. The Spoos just have to small of a range for a trading vehicle.

As far as the backtesting, I wanted to document the actual trades in real-time for a period of time to justify wether they were right or not.

The signals are right on the backtest results found here

To address the point of possibly making the trades longer. I refuse to hold futures overnight for I think the Risk/Reward will never be worth it. JMHO

You can take the result posted in the link and subtract 25% for slippage if you like. After watching the actual trades in real-time and then comparing them to the backtest results. I would say that 25% slippage would way over compensate for any slippage. 5% to 7% percent slippage would be a reasonable and more likely result.

Anyways it was reassuring to be able to match real-time results with the backtest.

[FX989231]

FibbGann

What happend after the last post?

I am really interested to know if the system is complete.

Regards

Kobus

[fulcrumX]

automated system

Hi

Although I have used esignal for some three years, this is my first ever post.

A big hello to everyone

I have developed a series of trading strategies which give me very good results.

Although not every trade is profitable, all of my strategies pick the correct direction.

I use momentum to identify if the trades will yield a few ticks or bigger yields.

What I need is someone to help me create an automated alert system that will signal a trade.

Any body up for it?