The EA wont even attach to the chart. How can I test anything.

Anyone having the same problems or is it just me?

It's ok.

I compiled it with no problem and did backtest as well.
No problem.
This EA is using some files from /include folder: stdlib.mqh and Tracert.mqh.
May be you already have it.

May be this EA is having the bug but this EA should be forward tested during at least one week to understand about this bug.
But I think that this EA may be very profitable.
Anyway I think that it's just the first version.

Can you suggest why I am unable to attach the EA after compiling it also?
Where should I unzip the tracert.mqh before I compile the EA?

BTW I can see a file tracert.mqh in "library" folder

Are you getting any backtest results on historical data that you can post?

Thanks, looking forward to attach this to the chart and get around to tweaking it

Hi,
Look post #48 in this thread.

Igor

__________________
Let's improve trade skills together
http://finance.groups.yahoo.com/group/TrendLaboratory

This file should be in /include folder. Not library folder.
An other one: "include" folder.
It should be place to this folder unzipped already. Just a file.

Looking on calculations made in excel, in file attached by radicalmoses, I've noticed, that sometimes Exit Price is higher than High Price of the day. Why?

Attached Images

volatility.jpg (65.3 KB, 882 views)

Sorry guys,

I am a little slow when it comes to these kinda stuff, but finally managed to get the EA working.

I have done the following changes to parameters:

Buy/sell : 80%
Stops : 60%

Lots to trade : 1.0 ( from 0.1)

The results are pretty good . However over a period of 5 months there have been many days without any trade. But the results are about 60% returns in 5 months or about 12% per month.


I am posting here my results of the backtest if any one else can confirm the same.

Thanks and a big thanks to Igorad for taking time to making an EA.
Anyone else getting better results with different parameters, please let me know.

Attached Files

StrategyTester.htm (97.8 KB, 271 views)

I am sure there are descripencies in the calculations as they were done manually. But what we are interested is in backtesting this with an EA as this is very simple system and works.

If anyone has downloaded the data from alpari and can get a higher modelling quality for this backtest, we will get a better idea how this performs.

So far I have done various combinations and it looks like 80%/60% works pretty good. I will continue testing this further.

Please change lots to 1.0.

A further test result with the following parameter changes. It just gets better.

Buy/Sell : 130%

Stop : 70%

Lots : 2 (previous test it was 1)

The win percentage is gone up to 70% and profit factor is about 2.5 which I think is excellent.

Consecutive losses is only 1 while consecutive wins is 3.

This looks quite profitable and we just might have a great set and forget system.

I hope some one can back me up with a better test result as I need confirmation with higher modelling quality.

I am getting optimistic.Imagine a totally mechanical system without any indicators?

Thanks guys, I hope the waiting was worth it.
A special thanks to Igorad for taking the trouble.

Attached Images

StrategyTester1.2.gif (5.1 KB, 715 views)

Attached Files

StrategyTester1.2.htm (95.6 KB, 148 views)

Hi,

I'm really glad to know that EA is working
For changing of lots number is good idea to use L.Williams Money Management (see Chapter 13 in his book).
For better results try to use TDW filter (Trade Day of Week) .
This EA has all these options.

Igor

__________________
Let's improve trade skills together
http://finance.groups.yahoo.com/group/TrendLaboratory