Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Nanningbob 4h trading system ver. 5.0 4,432 replies

Nanningbob 60x90 EA 3,920 replies

Help regarding NanningBob EA 27 replies

Nanningbob 4H trading system 2,017 replies

Help to improve profitable EA - 1H NanningBob 40MA 200 BB 10 replies

Attachments: Nanningbob V5: Steve Hopwood's auto-trading robot
Exit Attachments

Nanningbob V5: Steve Hopwood's auto-trading robot

Quoting Beastie Boy
Thanks for looking. In the current code, if the calculated values of BbSeparation or 1-BbSeparation = 0 (or are less than 0.1), then TradeDirection is set to none, ie we stop looking for a trade.

Please forgive me if I have this muddled up. I have read through it so many times I can no longer think straight!

Cheers, Beastie.
@ Beastie and f451, have you guys figured out what needs to be changed in the EA?
I'm confused Please sum up so we can fix the logic before week open, thanks guys.
24 hours in a day – 24 bottles of beer in a case – Coincidence? I think not
Quoting matrixebiz
@ Beastie and f451, have you guys figured out what needs to be changed in the EA?
I think the current code may be wrong, but confirmation is required from Steve or Simon. In a previous post, f451 quoted this line:

if ( (StochWhite > 85) && ( StochBlue > 85) && (MacdVal > OverBought) && (BbUpper > TopGreenLine) &&
((1-BbSeparation) < MinSepFromBB) )

which uses BbSeparation in the correct context and would allow a short trade at or close to the upper BB. However, the current code in post 1 is different as per my last post. It looks as thought the logic has been reversed such that instead of ((1-BbSeparation) < MinSepFromBB) confirming trading conditions are correct, it now sets TradeDirection to False, thus blocking the trade.

If I am correct, it is a simple fix and I am sure Steve or Simon will post an update soon.

Cheers, Beastie.
Hey guys,

I have had both EA's on my instaforex (I know, crappy broker, but I wanted to find out how bad with my 100.00) and it is taking no trades.

This is NANO, so is there a special setting here for either EA? Dazed and Confused....

Ok,

This is strange, after I wrote this, the Max version just fired off a buy stop!

Make that 4 orders! hmmmm.... I guess rebooting it was required? Strange.
We are our own best indicator.
Are you seeing setups appear that the EA should be taking, but isn't? Is anyone getting any #1 or #2 trades with the latest code? If not, then it could be due to the possible error that I refer to above.

If this is the case, and we still haven't heard from Steve or Simon, then I will post a modified version tomorrow. I don't really feel comfortable posting altered versions of someone elses work but Steve did say he would not be available over the weekend and I know you will all want a working version before next weeks kick off.

Unfortunately I can't get to my PC at the moment and so can't run any backtests. It may be useful for someone to run a test using the current code with only Allow1 and Allow2 enabled, Allow3 and Allow4 set to false.

Cheers, Beastie.
Quoting Beastie Boy
Are you seeing setups appear that the EA should be taking, but isn't? Is anyone getting any #1 or #2 trades with the latest code? If not, then it could be due to the possible error that I refer to above.

If this is the case, and we still haven't heard from Steve or Simon, then I will post a modified version tomorrow. I don't really feel comfortable posting altered versions of someone elses work but Steve did say he would not be available over the weekend and I know you will all want a working version before next weeks kick off.

Unfortunately...
Just downloaded latest steve version from post 1 and did a quick backtest for this month with only Allow1 and Allow2 and it did place trades
Attached Image (click to enlarge)
Click to Enlarge Name: StrategyTester.gif Size: 5 KB
24 hours in a day – 24 bottles of beer in a case – Coincidence? I think not
Quoting matrixebiz
Just downloaded latest steve version from post 1 and did a quick backtest for this month with only Allow1 and Allow2 and it did place trades
Thanks for that. I'm glad to see it still works, but now I am really confused. Would you be so kind as to run the same test on the attached version and compare the number of trades taken?
Please note everyone that I post this purely for testing and it is not sanctioned by Steve.

Cheers, Beastie.

PS. Good backtest results

EDIT - Attachment removed as it was for testing only.
Quoting Beastie Boy
Thanks for that. I'm glad to see it still works, but now I am really confused. Would you be so kind as to run the same test on the attached version and compare the number of trades taken?
Please note everyone that I post this purely for testing and it is not sanctioned by Steve.

Cheers, Beastie.

PS. Good backtest results
15 trades with post1 version and 11 trades with this version, mind you I'm not testing this on H4
24 hours in a day – 24 bottles of beer in a case – Coincidence? I think not
Quoting matrixebiz
15 trades with post1 version and 11 trades with this version, mind you I'm not testing this on H4
I know I'm asking a lot, but could you run both tests again with just Allow2 enabled? A quick look at the code made me realise that all your trades form the first test may have been #1.

Eternally grateful, Beastie.
Quoting Beastie Boy
I know I'm asking a lot, but could you run both tests again with just Allow2 enabled? A quick look at the code made me realise that all your trades form the first test may have been #1.

Eternally grateful, Beastie.
No worries

ok with only Allow2 - 0 trades with post1 version and 3 trades with your version

I'll have to check the version I'm using at home because I have had #2 trades last week
24 hours in a day – 24 bottles of beer in a case – Coincidence? I think not
matrixebiz, you are a star. It would appear that the post 1 version is blocking #2 trades, but allowing more #1 trades, although the #1 trades taken with the post 1 version may not be the same ones taken with the modded version.

This tells me that this Bbseparation check is not required (and shouldn't be used) for #1 and #2 trades.
#4 trades don't need this check since they will only trigger if the price is within Mode4DistanceToBB (5 pips) if the BB.
#3 trades do require this check to ensure that trades are not taken far away from the BB.

I will make some ammendments to what we currently have and post an updated version tomorrow, unless super speedy Steve beats me to it

Thanks again matrixebiz

Cheers, Beastie.
It was only a simple fix, so here it is.

PLEASE NOTE THAT THE ONLY OFFICIAL STEVEHOPWOOD SANCTIONED VERSION IS IN POST 1 OF THIS THREAD. THIS IS A MOD BASED ON MY OWN CONCLUSIONS BUT I DO BELIEVE THEY ARE SOUND.

The second disclaimer is that I was forced to make these changes in WordPad and so it has not been checked for compilation errors. If there are any, please shout.

Steve, I don't want to tread on anyone's toes. I will remove this if required.

With that out of the way....
Attached File(s)
File Type: mq4 Nanningbob V5 auto-trader by Steve and Simon #3 BbSeparation.mq4   87 KB | 192 downloads
For completeness, here are the modified pieces:-

Inserted Code
 //if ((1-BbSeparation) < MinSepFromBB) TradeDirection = none;

Inserted Code
if (Allow3 && (1-BbSeparation) < MinSepFromBB)
         {
            if (BbUpper >= BulgeTop && Bulge(TradeDirection,BbUpper,BbLower))
            {
               TradingStatus = mode3sell;
               GlobalVariableSet(GvName, 33);         
            }//if (BbUpper >= BulgeTop)
         }//if (Allow3)

Inserted Code
//if (BbSeparation < MinSepFromBB) TradeDirection = none;

Inserted Code
if (Allow3 && BbSeparation < MinSepFromBB)
         {
            if (BbLower <= BulgeBottom && Bulge(TradeDirection,BbUpper,BbLower)) 
            {
               TradingStatus = mode3buy;
               GlobalVariableSet(GvName, 3);
            }//if (BbLower <= BulgeBottom) 
         }//if (Allow3)
Quoting Beastie Boy
It was only a simple fix, so here it is.

PLEASE NOTE THAT THE ONLY OFFICIAL STEVEHOPWOOD SANCTIONED VERSION IS IN POST 1 OF THIS THREAD. THIS IS A MOD BASED ON MY OWN CONCLUSIONS BUT I DO BELIEVE THEY ARE SOUND.

The second disclaimer is that I was forced to make these changes in WordPad and so it has not been checked for compilation errors. If there are any, please shout.

Steve, I don't want to tread on anyone's toes. I will remove this if required.

With that out of the way....
Compiles fine, great work.
Do you think we still need to change f451's SetTarget(string tradedirection) section as he mentioned in a previous post;
http://www.forexfactory.com/showpost...postcount=1039
24 hours in a day – 24 bottles of beer in a case – Coincidence? I think not
Quoting matrixebiz
Do you think we still need to change f451's SetTarget(string tradedirection) section as he mentioned in a previous post;
http://www.forexfactory.com/showpost...postcount=1039
I like the code posted there by f451. It is a very elegant solution for profit target. I'm sure Steve will include it, but that's his call.

Cheers, Beastie.
I've backtested your version, Beastie boy. I backtested from 2007 onwards. Results are not amazing...

I tried default settings, then i tried playing around with the zone A and B and also with just using #1 and #2 methods.

What time frame did you test?
Quoting Viv108
I've backtested your version, Beastie boy. I backtested from 2007 onwards. Results are not amazing...

I tried default settings, then i tried playing around with the zone A and B and also with just using #1 and #2 methods.

What time frame did you test?
I now, it's not Bob's time frame... but try it on H1... nice results.
And don't forget that the trades are managed by MPTM... gives different results to only the EA!
"The only successful substitute for brains is silence."
Quoting DXTrade
And don't forget that the trades are managed by MPTM... gives different results to only the EA!
This is an important point. The purpose of Steve's EA is to try and make entries as per nanningbob's rules. Anyone trading the system manually will be monitoring open trades and taking profit if the trade starts to turn against them. Because the robot is automatic, some winning positions may be allowed to go into recovery. Although these trades may end as a break even, there is a lot of time wasted when profit could be being made.
Running this system in forward test, with MPTM managing trades, should yield different results. A move to break even when in profit, followed by a trailing stop should allow the system to catch some nice moves.

With regards to time frame, please be aware that some of the indicators are internally coded to work on the H4 time frame, regardless of the timeframe on the chart.
To use the EA properly on other time frames, open the mq4 file in MetaEditor, and do a Find and Replace. Searh for PERIOD_H4 and replace it with 0 (that is a zero).

It will be interesting to see what results people get on different time frames. I have even had some excellent results on M1! I tested this as a means to try and reduce draw down, my thinking being that on a lower time frame, ReEntryLinePips could be set to a lower value so that recovery starts sooner.

This will be a good debate once we are happy that the EA is taking the correct trades.

Cheers, Beastie.
Steve, in the current code, WindowPriceMax and WindowPriceMin are used to determine screentop and screenbottom. This causes a problem in that users will get different results based on their zoom level on the chart. This is OK in Bobs manual strategy, as it is done by eye, but as this is automatic, it would be better if WindowPriceMax and WindowPriceMin were calculated.

I have found that the values can be derived by taking the range of prices from lowest low to highest high, and adding 5% to find the screentop and subtracting 2.4% to find the screenbottom.

If you think this is reasonable, I have pasted code below.

Inserted Code
//Simon's #3 Bulge variable
   //double screentop = WindowPriceMax(0);   
   //double screenbottom = WindowPriceMin(0);
   double range = High[iHighest(NULL,0,MODE_HIGH,BarCount,1)] - Low[iLowest(NULL,0,MODE_LOW,BarCount,1)]; 
   double screentop = High[iHighest(NULL,0,MODE_HIGH,BarCount,1)] + (range / 100 * 5); 
   double screenbottom = Low[iLowest(NULL,0,MODE_LOW,BarCount,1)] - (range / 100 * 2.4);    
   double value = screentop - screenbottom;      //value top of the chart - value bottom
   double sixth = value/6;
   double valueS = (value*(MathPow(10,Digits)));
   double sixthS = (sixth*(MathPow(10,Digits)));
Cheers, Beastie.
yes you are right
My bot is coding to work with M15, H1 or H4 Tf
so far I have the best results with TH = H1
the robot have now 3 strategies :
- Nb
- simon trend
- Trix (idea from THV), so far only in testing mode, draw arrows, not take any trades

Now this bot and master Ea have more than 8000 coding lines (Yes it's an increadible monster, result of my 20 years trading, 4 years coding, and other outstanding coders like Steve, Simon ect...), so i will not realease mq4 any more, only ex4, anyway I think nobody can understand it.
Play with it (best on H1 tf, also not bad on M15, rather poor on H4) and it swallow pips on an outstanding manner (so far)
I hope people follow my boy have all the indics
It's really a monster that will kill your cpu if you play it on many pairs (I play 26 pairs), althought i use some feature in order to calculate some functions not every ticks
Attached File(s)
File Type: ex4 NBv5-mtf-mx075ff.ex4   120 KB | 210 downloads
File Type: ex4 Master EA_008.ex4   54 KB | 198 downloads
Quoting Beastie Boy
Steve, in the current code, WindowPriceMax and WindowPriceMin are used to determine screentop and screenbottom. This causes a problem in that users will get different results based on their zoom level on the chart. This is OK in Bobs manual strategy, as it is done by eye, but as this is automatic, it would be better if WindowPriceMax and WindowPriceMin were calculated.

I have found that the values can be derived by taking the range of prices from lowest low to highest high, and adding 5% to find the screentop and subtracting 2.4% to find the screenbottom....
very well spotted, i have the same problem when i zoom, i will study your code

Cheers

Max
0 traders viewing now