High & Low 8 Hours (GBP/JPY) | Page 113 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

What are Higher High, Higher Low, Lower High and Lower Low? 11 replies

ECN && STP && Scalping && Hedging - Broker Review - Oct 2009 24 replies

Gbp/Jpy High Low Trading System 2 replies

High to High and Low to low time cycles 1 reply

GBP/JPY high or low? 4 replies

Attachments: High & Low 8 Hours (GBP/JPY)
Exit Attachments

High & Low 8 Hours (GBP/JPY)

  • #2,241
  • Jan 16, 2008 8:54am Jan 16, 2008 8:54am
Quoting harwedar
I want to Congratulate, you people for bumper harvest of pips you reap with this system. Tks the originator, the optimizers, the coder and the entire forum.
Now, I want to ask you people jublitating 100s of pip, did you use the ea or you manual trade the strategy.

thanks.

harwedar.
It's the ea for me. I am tucked up in bed when the manual orders need placing. I could not trade this strategy without the ea.

  • #2,242
  • Jan 16, 2008 8:58am Jan 16, 2008 8:58am
Quoting harwedar
I want to Congratulate, you people for bumper harvest of pips you reap with this system. Tks the originator, the optimizers, the coder and the entire forum.
Now, I want to ask you people jublitating 100s of pip, did you use the ea or you manual trade the strategy.

thanks.

harwedar.
I do it manually. Its 6 in the morning for me which is fine also I'm not good at ea's.
BK
  • #2,243
  • Jan 16, 2008 9:37am Jan 16, 2008 9:37am
Quoting saaedb
I've been running the EA since day one and I've not seen that error on my live or demo account. So its not very easy for me to be able to figure out the issue. But my guess is, the EA tries to read your history tab and fails to read it and opens the new orders. Make sure you have set your history tab to show all account history.
I'll see if there is anything I can do on my part to prevent this.

Saaedb
Saaedb - it did it to me again this morning on GJ for a $75.45 loss right after closing the TP. The open and close time were the same as the close time of the TP. I am now using your latest EA but it does the same as the previous. I have my buffer set to 8 and resets set to 1. The rest is normal. Gerrigirl is having the same prob. Can you post the EA again and I will download and replace my existing. I don't think it will help. I also trade EJ and GU, but haven't seen the problem. I have attached my live snapshot.

[IMG]file:///C:/DOCUME%7E1/ADMINI%7E1/LOCALS%7E1/Temp/moz-screenshot.jpg[/IMG][IMG]file:///C:/DOCUME%7E1/ADMINI%7E1/LOCALS%7E1/Temp/moz-screenshot-1.jpg[/IMG][IMG]file:///C:/DOCUME%7E1/ADMINI%7E1/LOCALS%7E1/Temp/moz-screenshot-2.jpg[/IMG]
Attached File(s)
File Type: doc Extra GJ 16Jan08.doc   60 KB | 466 downloads
  • #2,244
  • Jan 16, 2008 9:54am Jan 16, 2008 9:54am
Quoting SteveHopwood
Done, and attached.

Thanks for the latest version, saaedb.

Steve -Thanks for updating my Word.doc. I haven't had much time.

Saaedb - I should have read to the end - thanks for the updated EA - hope that stops the extra trade on a GJ TP.
  • #2,245
  • Jan 16, 2008 9:59am Jan 16, 2008 9:59am
Quoting LarryR
Saaedb - it did it to me again this morning on GJ for a $75.45 loss right after closing the TP. The open and close time were the same as the close time of the TP. I am now using your latest EA but it does the same as the previous. I have my buffer set to 8 and resets set to 1. The rest is normal. Gerrigirl is having the same prob. Can you post the EA again and I will download and replace my existing. I don't think it will help. I also trade EJ and GU, but haven't seen the problem. I have attached my live snapshot.

[IMG]file:///C:/DOCUME%7E1/ADMINI%7E1/LOCALS%7E1/Temp/moz-screenshot.jpg[/IMG][IMG]file:///C:/DOCUME%7E1/ADMINI%7E1/LOCALS%7E1/Temp/moz-screenshot-1.jpg[/IMG][IMG]file:///C:/DOCUME%7E1/ADMINI%7E1/LOCALS%7E1/Temp/moz-screenshot-2.jpg[/IMG]
Did you have the same issue with the new EA? If so, like I said, I can't see why it would do this but again, I'll go in the code and put a few more checks into it. Which broker are you using? The only reason that I can see why the EA would do such a thing is, it can't read the history tab and thinks no trade has been opened for the day, or, it reads the history tab before the TP trade is reported in the history tab.

Here is one way to get around this problem, even though the EA will still open the second trade but it wont cause you a loss. Set the "AloowLimitOrders" variable to true. Like this when the EA does open a second trade it will open it as a limit order and remove it, rather then a market order and close it at a loss.
Now on the flip side, if you attach the EA too late to the charts and the price has already crossed the high/low levels, it wont open a market order but instead a limit order at the original levels. which i think is good.

Please tell this to GarriGirl as wel..

Let me know how it works out for you.
Saaedb
  • #2,246
  • Jan 16, 2008 11:40am Jan 16, 2008 11:40am
Saaedb - I did not use the new EA last night. I have downloaded it and will try it tonight. I will update my word.doc to reflect the current EA pictures to match Steve's new words, as well. When I am up and try to outguess the trade - I usually guess wrong. Stay with the EA folks.

Thanks for your continued effort. It has made me money and I can only thank you. When you get to Canada, you have a place to stay and I will buy you several good Canadian brew. Your $ will go further here as well. http://www.forexfactory.com/images/icons/icon7.gif
  • #2,247
  • Jan 16, 2008 11:46am Jan 16, 2008 11:46am
I use IBFX. The history tab sounds right, because I sometimes don't see my closed trade for a few minutes. I will set LimitOrders to true and amend word.doc accordingly I will leave it as a word doc so others can tailor their own guidelines.
  • #2,248
  • Jan 16, 2008 12:00pm Jan 16, 2008 12:00pm
Quoting damonl
Hmmmm....What is your version of the system? I saw your post a few days ago and it seams that you made a profit every year though some were small.

Would you mind sharing how your settings are different than what is posted on the .doc file.

Thank you,

Damon
I now really don't know what to say. The version I described earlier is sitting here: http://www.forexfactory.com/showpost...postcount=2194. However, it was tested using a different programming approach, a different programming language and much cruder data than used by Saaedb's EA.

Saaedb did great work in creating what is almost "the ultimate breakout strategy tester" (kudos again and again , saaedb). I played with it today. I came up with what seem to be the optimal settings for 2007:

SL: 95
TP: 999
Entry Buffer: 1
RangeStart: 18 [GMT]
RangeEnd: 3 [GMT]
NumberOfResets: 2
MaxRange: 300
MinRange: 0
AllowLimitOrders:F
OnMondayUseWeekendDataOnly: T
CloseEndOfDayCandle: -- / 25
EnterOnRetracement: F
RetraceAfterBreakOut: --
Stop_LookingForTrades: 14 [GMT]
Hours_CloseTradeAfterOpen: 84
FixedLot: --
RISK: 4%
MoneyManagement: T
MinLotSizeAllowed: --
IsAccountMini: F
TrailingStop: 70
BreakEvenStop: --

These settings made fantastic ROI in 2007, about 500%... As far as we can trust backtesting...

HOWEVER, in 2006 they made about 10% only, and in the years before that, they simply swallowed all of the deposit!!!

I tweaked these setting to make them work for the previous years. I came up with something more similar to what I've posted previously:


SL: 125
TP: 125
Entry Buffer: 1
RangeStart: 16 [GMT]
RangeEnd: 7 [GMT]
NumberOfResets: 1
MaxRange: 250
MinRange: 50
AllowLimitOrders:F
OnMondayUseWeekendDataOnly: T
CloseEndOfDayCandle: -- / 25
EnterOnRetracement: T
RetraceAfterBreakOut: 15
Stop_LookingForTrades: 14 [GMT]
Hours_CloseTradeAfterOpen: 84
FixedLot: --
RISK: 4%
MoneyManagement: T
MinLotSizeAllowed: --
IsAccountMini: F
TrailingStop: --
[FONT='Times New Roman','serif']BreakEvenStop[/font][FONT='Times New Roman','serif']: -- [/font]

This brought the ROI in 2007 for a "modest" 100%, HOWEVER, in the previous yeras, these settings would still loose money (not as much as the other version) or BE.

Add to all this the fact that in its orifginal form, this system would make very good money in 2007, but loose all of the depo in previous years.

So what is the bottom line? I think it can't be expressed better than your:

Quoting damonl
---It also could be that these more volitile times will be here for a while which would be good for almost any system that trades GJ.
---
I'm not sure that it's exactly the volatility, but there was something special about 2007 that made almost every GJ system perform well... So, "rejoice, o young man, in thy youth", but remember that an end will come to your joy - and be ready to identify this moment...

Good luck to all.
sans peur et sans reproche
  • #2,249
  • Jan 16, 2008 2:30pm Jan 16, 2008 2:30pm
Quoting LarryR
Saaedb - I did not use the new EA last night. I have downloaded it and will try it tonight. I will update my word.doc to reflect the current EA pictures to match Steve's new words, as well. When I am up and try to outguess the trade - I usually guess wrong. Stay with the EA folks.

Thanks for your continued effort. It has made me money and I can only thank you. When you get to Canada, you have a place to stay and I will buy you several good Canadian brew. Your $ will go further here as well. http://www.forexfactory.com/images/icons/icon7.gif
Thanks, I might have to take you up on this offer one day .
  • #2,250
  • Jan 16, 2008 2:33pm Jan 16, 2008 2:33pm
"the ultimate breakout strategy tester"

Its not the ultimate version yet. I have plans for this EA, I'm just waiting for the right time.
With the modification I have in mind, there will be alot more options to filter out trades and use multiple time frames. A person would be able to use this EA as a scalper and a intra day trader. But I wont be making that version of the EA public at least that is the plan as of now.
  • #2,251
  • Jan 16, 2008 3:59pm Jan 16, 2008 3:59pm
Quoting saaedb
"the ultimate breakout strategy tester"

Its not the ultimate version yet. I have plans for this EA, I'm just waiting for the right time.
With the modification I have in mind, there will be alot more options to filter out trades and use multiple time frames. A person would be able to use this EA as a scalper and a intra day trader. But I wont be making that version of the EA public at least that is the plan as of now.
I continued playing with this EA. It's possible to find, quite easily, a set of values for each parameter such that the breakout from a range formed during end of NY session - beginning of the Asian session would be profitable - in 2007. Other years are much more problematic.

I still have to work on it a lot (as the EA in its present form doesn't allow it), but the general idea for improving the system further is similar to the one that has been posted here (sorry, I can't find now, by whom): opening the trades only in the direction of the current main trend. After all, the trend IS your biggest (and only) friend.

Saaedb, I don't know whether it fits your plans and how much work it takes, but the EA would, in my eyes, be perfect/ultimate if you could add to it a possibility of filtering out trades using a MA: opening only long positions if the slope is ascending, opening only short positions if the slope is descending. If I may suggest, the attached MA is the best available, I rely on it heavily in my trading.

PS Actually, one more feature could do good to the EA. At present, there is only the possibility to filter out trades on Mondays and Fridays. From what I saw till now, Thursdays are the worst days for GJ. For other pairs, there could be other days (for example, I know that, frequently, Wednesdays are, for some reason, bad for GU). Could you consider adding the possibility of filtering out other days?
Attached File(s)
File Type: mq4 NonLagMA_v7.1.mq4   6 KB | 543 downloads
sans peur et sans reproche
  • #2,252
  • Jan 16, 2008 6:07pm Jan 16, 2008 6:07pm
Quoting saaedb
Did you have the same issue with the new EA? If so, like I said, I can't see why it would do this but again, I'll go in the code and put a few more checks into it. Which broker are you using? The only reason that I can see why the EA would do such a thing is, it can't read the history tab and thinks no trade has been opened for the day, or, it reads the history tab before the TP trade is reported in the history tab.

Here is one way to get around this problem, even though the EA will still open the second trade but it wont cause you a loss. Set the "AloowLimitOrders" variable to true. Like this when the EA does open a second trade it will open it as a limit order and remove it, rather then a market order and close it at a loss.
Now on the flip side, if you attach the EA too late to the charts and the price has already crossed the high/low levels, it wont open a market order but instead a limit order at the original levels. which i think is good.

Please tell this to GarriGirl as wel..

Let me know how it works out for you.
Saaedb
Thank You for trying to fix the extra order problem. I am going to try the new ea tonight. You may be right about it having to do with the history tab. I remember one time that it happened after I was sorting the history to show a custom period. But the other times I just don't know.
Thanks again, hope your having a great vacation!
My orders tonight
Buy 212.09
Sell 209.62
  • #2,253
  • Jan 16, 2008 6:24pm Jan 16, 2008 6:24pm
Quoting ilanr
I now really don't know what to say. The version I described earlier is sitting here: http://www.forexfactory.com/showpost...postcount=2194. However, it was tested using a different programming approach, a different programming language and much cruder data than used by Saaedb's EA.

Saaedb did great work in creating what is almost "the ultimate breakout strategy tester" (kudos again and again , saaedb). I played with it today. I came up with what seem to be the optimal settings for 2007:

SL: 95
TP: 999
Entry Buffer: 1
RangeStart: 18 [GMT]
RangeEnd: 3 [GMT]
NumberOfResets: 2
MaxRange: 300
MinRange: 0
AllowLimitOrders:F
OnMondayUseWeekendDataOnly: T
CloseEndOfDayCandle: -- / 25
EnterOnRetracement: F
RetraceAfterBreakOut: --
Stop_LookingForTrades: 14 [GMT]
Hours_CloseTradeAfterOpen: 84
FixedLot: --
RISK: 4%
MoneyManagement: T
MinLotSizeAllowed: --
IsAccountMini: F
TrailingStop: 70
BreakEvenStop: --

These settings made fantastic ROI in 2007, about 500%... As far as we can trust backtesting...

HOWEVER, in 2006 they made about 10% only, and in the years before that, they simply swallowed all of the deposit!!!

I tweaked these setting to make them work for the previous years. I came up with something more similar to what I've posted previously:


SL: 125
TP: 125
Entry Buffer: 1
RangeStart: 16 [GMT]
RangeEnd: 7 [GMT]
NumberOfResets: 1
MaxRange: 250
MinRange: 50
AllowLimitOrders:F
OnMondayUseWeekendDataOnly: T
CloseEndOfDayCandle: -- / 25
EnterOnRetracement: T
RetraceAfterBreakOut: 15
Stop_LookingForTrades: 14 [GMT]
Hours_CloseTradeAfterOpen: 84
FixedLot: --
RISK: 4%
MoneyManagement: T
MinLotSizeAllowed: --
IsAccountMini: F
TrailingStop: --
[FONT='Times New Roman','serif']BreakEvenStop[/font][FONT='Times New Roman','serif']: -- [/font]

This brought the ROI in 2007 for a "modest" 100%, HOWEVER, in the previous yeras, these settings would still loose money (not as much as the other version) or BE.

Add to all this the fact that in its orifginal form, this system would make very good money in 2007, but loose all of the depo in previous years.

So what is the bottom line? I think it can't be expressed better than your:



I'm not sure that it's exactly the volatility, but there was something special about 2007 that made almost every GJ system perform well... So, "rejoice, o young man, in thy youth", but remember that an end will come to your joy - and be ready to identify this moment...

Good luck to all.
Which data are you using to do the back tests and what is your modelling quality?
  • #2,254
  • Jan 16, 2008 6:29pm Jan 16, 2008 6:29pm
Here are my orders for today: VTtrader
HI: 211,97
LOW: 209,62
  • #2,255
  • Jan 16, 2008 10:50pm Jan 16, 2008 10:50pm
Quoting ilanr
I don't think it could be the explanation. Indeed, GJ was very volatile in 2007 (average daily range of 229 p.) and the system worked well, but 2006 had a very low volatility for GJ (154 p.) and still it was a good year for this system (at least my version of it). 2004 and 2000 had the same volatilty (204 p), but the first was good and the second was bad.
I appreciate this. If movement was acceptable for the trade even in the bad years, the only thing it could be is that the time the movement occurred was different then it is today.

Obviously in those bad years, if we entered the market at the same time we are now, it was just slow then and we got stopped out.

I bet if you changed the time frame in the bad years, you would have good results.

Maybe this has already been mentioned.

So then the real question is how would you have known to switch times?
"Keep your eyes on the helpers" - Mr. Rogers
  • #2,256
  • Jan 16, 2008 11:33pm Jan 16, 2008 11:33pm
Quoting saaedb
Which data are you using to do the back tests and what is your modelling quality?
I used MetaQuotes M1 data, modelling quality 90%.
sans peur et sans reproche
  • #2,257
  • Jan 16, 2008 11:47pm Jan 16, 2008 11:47pm
Saaedb,

thanks for including the limit orders option, but is it possible to use this to set a limit order a desired pip distance away from the box, ie. if price penetrates the original buy value, to allow for a retrace then entry on a limit order??????
TRADE FOR SWAP
  • #2,258
  • Jan 17, 2008 1:02am Jan 17, 2008 1:02am
good luck guys...
  • #2,259
  • Jan 17, 2008 3:16am Jan 17, 2008 3:16am
Quoting cgldsmth
Saaedb,

thanks for including the limit orders option, but is it possible to use this to set a limit order a desired pip distance away from the box, ie. if price penetrates the original buy value, to allow for a retrace then entry on a limit order??????
yes, go back a few pages and read my explanation where I posted the last EA.
  • #2,260
  • Jan 17, 2008 3:34am Jan 17, 2008 3:34am
Quoting ilanr
I used MetaQuotes M1 data, modelling quality 90%.
http://66.179.204.176/gbpjpy/Strateg..._7_Percent.htm
0 traders viewing now