stevehopwoodforex.com
https://www.stevehopwoodforex.com/phpBB3/
Print view

Bears'n'Bulls EA - V1.05g - in forward- & live-test
https://www.stevehopwoodforex.com/phpBB3/viewtopic.php?t=1373		 Page 13 of 16
Author:  traveliron [ Thu Apr 18, 2013 2:04 pm ]
Post subject:  Congratulations Banzak
Hi Banzak

In your last post you said "At the moment I'm working at the vesion V1.05a and I will post it here as soon as I have some good results." Well, you already have some results that are better than "good". I don't quite know how I've missed this until now but BIG congratulations on an outstanding EA.

Looking at the results on your Digitrade account it appears you have overcome the quite large DD that was a feature of the previous examples. Is this a result of a change in broker, optimised settings or manual intervention? Whatever it is that's an amazing result! I know it's still too early for the more discerning members of this community but those results cannot be ignored.

It seems that EA has been slowly but successfully chugging away whilst you've been working on further developments. Are you any closer to the v1.05a release?

I assume it's V1.04a so can you post the .set you used on the Digitrade account and tell me if there are any other adjustments or optimisations you'd like me to forward test. I plan to use Global Prime and get the results posted up on this forum.

Once again, congratulations! Outstanding work.

Regards - Chris
Author:  banzak [ Thu Apr 18, 2013 4:06 pm ]
Post subject:  Re: Congratulations Banzak
traveliron wrote:Hi Banzak

In your last post you said "At the moment I'm working at the vesion V1.05a and I will post it here as soon as I have some good results." Well, you already have some results that are better than "good". I don't quite know how I've missed this until now but BIG congratulations on an outstanding EA.

Looking at the results on your Digitrade account it appears you have overcome the quite large DD that was a feature of the previous examples. Is this a result of a change in broker, optimised settings or manual intervention? Whatever it is that's an amazing result! I know it's still too early for the more discerning members of this community but those results cannot be ignored.

It seems that EA has been slowly but successfully chugging away whilst you've been working on further developments. Are you any closer to the v1.05a release?

I assume it's V1.04a so can you post the .set you used on the Digitrade account and tell me if there are any other adjustments or optimisations you'd like me to forward test. I plan to use Global Prime and get the results posted up on this forum.

Once again, congratulations! Outstanding work.

Regards - Chris
Hi traveliron,
i'm still on debugging with the V1.05a, cause there are still some things I don't like in the trading behavior. The low DD at MydigiTrade is cause I intervene there manually, when I see that the trend is changing and the decision of the EA is not correct. I know when my EA makes the wrong decision, this is the reason why I'm developing the V1.05a. :)

Would be great if you do some forward test on Global Prime, I also opened an account today.

If you need some help, contact me. I advise you should do a Backtest before you go live on Global Prime, for sure the setfile has to be adjusted and tested. If you don't have much experience in backtesting, i will do it for you.

Regards,
Banzak
Author:  traveliron [ Thu Apr 18, 2013 9:21 pm ]
Post subject:  Getting Ready
Ah! I thought the low DD might be might be something to do with manual intervention. Is there a technical indication you are using to identify the trend change or are you just looking at the chart and seeing the bleeding obvious with your eyeballs? Whatever technique you're using it's very effective. It would be fantastic to have the EA detect trend changes and correct itself. But it may be that you end up with essentially a semi-automatic system that periodically needs a little intervention. Like a toy train that sometimes needs to be put back on it's track. With the results you're demonstrating I could certainly live with that.

Unfortunately I'm very busy over the next few days but everything will ease off on Tuesday. I'll get set-up then and let you know if I have any problems.

I have never carried out backtesting before, there are a lot of wise people on these fora that scoff at the idea and think its next to useless. But I can see that it will almost certainly help optimise settings for your EA, so I'll take your advice and get on to that too. It is kind of you to offer, but if I have problems with backtesting I'll resolve it with the help of others. I want you to concentrate on what you do best - creating stonking EAs.

Thanks - Chris
Author:  SWG123 [ Thu Apr 18, 2013 10:00 pm ]
Post subject:  Re: Getting Ready
traveliron wrote:Ah! I thought the low DD might be might be something to do with manual intervention. Is there a technical indication you are using to identify the trend change or are you just looking at the chart and seeing the bleeding obvious with your eyeballs? Whatever technique you're using it's very effective. It would be fantastic to have the EA detect trend changes and correct itself. But it may be that you end up with essentially a semi-automatic system that periodically needs a little intervention. Like a toy train that sometimes needs to be put back on it's track. With the results you're demonstrating I could certainly live with that.

Unfortunately I'm very busy over the next few days but everything will ease off on Tuesday. I'll get set-up then and let you know if I have any problems.

I have never carried out backtesting before, there are a lot of wise people on these fora that scoff at the idea and think its next to useless. But I can see that it will almost certainly help optimise settings for your EA, so I'll take your advice and get on to that too. It is kind of you to offer, but if I have problems with backtesting I'll resolve it with the help of others. I want you to concentrate on what you do best - creating stonking EAs.

Thanks - Chris
Hi Chris,

IMO quality backtesting is essential but lMT4 using metaquotes history is not great. If you want to do it properly look here:-

http://eareview.net/tick-data
Author:  traveliron [ Fri Apr 19, 2013 8:17 am ]
Post subject:  Re: Bears'n'Bulls EA - actual version V1.04a - stable
Thanks SWG123

I'll have a good look at that and probably bother you later if I run into problems.

I used to be a muppet too; it was slang for the "Movements" trade in the RAF.
Author:  theforexedge [ Fri Apr 19, 2013 8:50 am ]
Post subject:  Re: Bears'n'Bulls EA - actual version V1.04a - stable
Travel Iron

Backtesting is a waste of time, but if you use Tick Data & Walk Forward Analysis, you can evaluate & turn a profitable strategy. Unfortunately backtesting cannot make a unprofitable strategy profitable, and most people end up curve fitting and then are surprised when there account is blown-up

http://www.easyexpertforex.com/walk-for ... rader.html

I recommend this book as well:

http://www.amazon.co.uk/Evaluation-Opti ... 0470128011

The big problem with Empty4 is that it doesn't support multicore processing, so its painfully slow.

Hope this helps..

Jason
Author:  SWG123 [ Fri Apr 19, 2013 5:07 pm ]
Post subject:  Re: Bears'n'Bulls EA - actual version V1.04a - stable
theforexedge wrote:
Backtesting is a waste of time

I recommend this book as well:

http://www.amazon.co.uk/Evaluation-Opti ... 0470128011

Jason
Hmm, Robert Pardo seems to disagree with you Jason:- :D

"The main reason that we test a trading strategy is to see whether it works, that is, produces a profit. Indeed, perhaps the first and foremost advantage of a thoroughly tested systematic trading strategy is the determination that it, in fact, has a profit potential. Another way of looking at this is that a successful and fully tested systematic trading strategy is in itself a proof of the trading concept. Without a reasonable estimate of the potential risk-adjusted reward of a trading strategy, it is impossible to know if it is worth trading. Without a reasonable estimate of potential risk, it is impossible to know the true cost
of trading with the strategy.

How is it that we determine that a trading strategy has a positive profit expectancy? This is done through the construction and evaluation of a historical simulation.

Why is it that the strategist goes through all of the work to painstakingly construct and evaluate a historical trading simulation just to verify the validity of a trading system? The answer to that question is quite simple and most basic. We want to see whether it works, that is, that it produces a trading profit on past historical market data."
Author:  theforexedge [ Fri Apr 19, 2013 6:10 pm ]
Post subject:  Re: Bears'n'Bulls EA - actual version V1.04a - stable
I wasn't disagreeing I was just pointing out if you follow his methods then back testing with out of sample data is effective...using the standard Empty4 back testing is a waste of time. sorry if my post was confusing!!
Author:  SWG123 [ Fri Apr 19, 2013 6:20 pm ]
Post subject:  Re: Bears'n'Bulls EA - actual version V1.04a - stable
theforexedge wrote:I wasn't disagreeing I was just pointing out if you follow his methods then back testing with out of sample data is effective...using the standard Empty4 back testing is a waste of time. sorry if my post was confusing!!
In that case we are on the same page - cheers! :D
Author:  sponn [ Wed Apr 24, 2013 12:46 pm ]
Post subject:  Re: Bears'n'Bulls EA - actual version V1.04a - stable
Hello Banzak...

Do you still need backtesters ? I can help a little with my home and vps. Just let me know which one pairs do you still need to be bt.

BTW Great bot !
All times are UTC Page 13 of 16

Powered by phpBB® Forum Software © phpBB Limited