Experts Tool Box | Page 38 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Box Spreads of Equity Index Options on Futures as a Financing Tool 0 replies

experts advisor file not in experts folder 12 replies

box play 18 replies

Anyone trading "Box Options" with Oanda??? 20 replies

Oanda Order Flow Box 5 replies

Attachments: Experts Tool Box
Exit Attachments

Experts Tool Box

  • #741
  • Apr 29, 2012 3:19pm Apr 29, 2012 3:19pm
Quoting Baffa
I'm relative new to this thread.

I'm testing the ETB v109 and as I understand you wont be working on this EA anymore.

I agree that the EA is too full. I don't use most of the tools as my trade is based on The trade made simple thread.

My concern is that you won't be able to please everyone and make a specific EA for everyone.
I must say I'm enjoin the ETB EA, and soon I'll be posting results of my tests. I really appreciate your work and effort and I'll be happy to help in any way I can.

I have a question generic to any EA. Do you have...
If your concern with spread is if it is larger than a preset value can the trade be aborted, the answer is yes if it is one of the filters in the EA.
  • #742
  • Apr 29, 2012 3:22pm Apr 29, 2012 3:22pm
Quoting Jd88
(sorry about the typo in your "handle" - been up for 18 hours - no dinner yet. You know the life!

Thanks Seller9 – your suggestion to start with a simpler clean slate to create the TDI EA is both logical and generous.

I have never done this before – so I hope it is useful to you. Please tell me what else you need and if I should send it to you PM / offline.

Not in any particular order.

My Manual Model for approx 2 years is largely a classic and accurate application of the original TDI method. After thousands of hours of experimentation...
I will take a look at your method and see what I can do.
  • #743
  • Apr 29, 2012 6:43pm Apr 29, 2012 6:43pm
Quoting Jd88
(sorry about the typo in your "handle" - been up for 18 hours - no dinner yet. You know the life!

Thanks Seller9 – your suggestion to start with a simpler clean slate to create the TDI EA is both logical and generous.

I have never done this before – so I hope it is useful to you. Please tell me what else you need and if I should send it to you PM / offline.

Not in any particular order.

My Manual Model for approx 2 years is largely a classic and accurate application of the original TDI method. After thousands of hours of experimentation...
If your method is so accurate, why do the ema settings have to be adustable?
  • #744
  • Apr 29, 2012 7:22pm Apr 29, 2012 7:22pm
Sorry I asked.

I took a look at your outline in the previous post and decided that it was not within the new rules of the forum of trying to "KISS". I thought you had said it was following the original TMS using TDI, but I see it has to many indicators and to many adjustments the user needs to make.

My idea is to make very simple EAs that can continue to grab a few pips from a group of pairs each day.

Your idea probably works from the live traders point of view as he can see at a glance after years of watching what is going to happen next. Putting it into code would probably contain as much code as the etb EA and would turn into a nightmare of settings by the trader.

So I am going to have to pass on this one.

Good Luck
  • #745
  • Apr 30, 2012 12:33am Apr 30, 2012 12:33am
Thanks for your time and for considering it.
  • #746
  • Edited 8:26am Apr 30, 2012 12:52am | Edited 8:26am
Quoting fxshadow
If your method is so accurate, why do the ema settings have to be adustable?
The adjustable settings I believed would allow the EA to come closer to mimicking the trade "human component (short of some sort of adaptive / genetic approach).

A simple on / off switch can work with certain EAs I guess. In most other cases though, I believed that an EA which can allow for "controls / adjustments" in managing the screens that were proven to work manually, variables can potentially be more effective.

My experience with EAs is limited - I purchased only 3 over the last 5 years - after reading review and opinion checking of existing users,. etc. (For the most part, I considered it just a crap shoot anyway.)
And when I tried these, they seem to make the same calls without allowing for enough of what i felt were necessary inputs. And they would also disregard what I felt were forces of significance in moving the market.

Again, as a non-coder and non-EA expert - my attempt was to provide a degree of control - by allowing an EA more adjust ability in the method components (indicators) that I proved to be effective and useful. Maybe my thinking is all wrong. It is really based on my manual trading experience - but that is clearly not enough.

I lost thousands and made thousands testing my manual methods - always with the hope that what I did could be automated one day. But I guess I am wrong. Anyway - my goal is to stay on a worthwhile learning track - and the route to sustained profitability.

And of course, I hope to stay connected to this thread and contribute in response to what Seller9 provides.

Sorry for the long-winded answer.
  • #747
  • Apr 30, 2012 1:31pm Apr 30, 2012 1:31pm
As I said privately to Seller9 my idea is that it's absolutely essential to fix the ea with a progressive average time filter. As I could see during my live tests, time is a fundamental variable using rb. Using 3-9-3, 3-13-3 but especially 2-2-12, 3-3-13 or similar, with a fixed acceptable "time for candle" < 180'' the vast majority of loosing position were avoided. But at the same time a considerable percentage of good entries were still lost. With a candle time back average we could achieve a more precise filter.
So if someone is capable to swiftly create such a code this could help Seller9 avoiding him some unnecessary loss of time.
Thank you for your attention
Ubi maior minor cessat
  • #748
  • Apr 30, 2012 2:56pm Apr 30, 2012 2:56pm
Here are some settings posted by Icemaan that you may want to try if your giving this EA a try.

Attached File(s)
File Type: txt Settings_Bar_BEJS.txt   1 KB | 644 downloads
  • #749
  • May 1, 2012 1:31pm May 1, 2012 1:31pm
Quoting seller9
If your concern with spread is if it is larger than a preset value can the trade be aborted, the answer is yes if it is one of the filters in the EA.

Hi Seller,

I didn't understand. Does the ETB EA has a filter for the spread? (if the spread is higher then "x" abort the trade)

If yes could you tell me the name of the input.

I'm sorry I'm still asking questions about this EA.

Thanks again

Baffa
  • #750
  • May 1, 2012 3:42pm May 1, 2012 3:42pm
Quoting Baffa
Hi Seller,

I didn't understand. Does the ETB EA has a filter for the spread? (if the spread is higher then "x" abort the trade)

If yes could you tell me the name of the input.

I'm sorry I'm still asking questions about this EA.

Thanks again

Baffa
No it doesn't, but that could be a filter in an EA, as it is easy to add.
  • #751
  • May 1, 2012 4:20pm May 1, 2012 4:20pm
Quoting seller9
No it doesn't, but that could be a filter in an EA, as it is easy to add.

Hi Seller,

It would be great to have that filter. I've been noticing that some trades have a good signal but with a huge spread and become losers.

Thanks again,

Baffa
  • #752
  • May 1, 2012 11:24pm May 1, 2012 11:24pm
Quoting Baffa
Hi Seller,

It would be great to have that filter. I've been noticing that some trades have a good signal but with a huge spread and become losers.

Thanks again,

Baffa
Its possible I may continue with the etb EA, but I need to clean out some stuff that is not being used to make room, the Max spread is real easy to add so will put it on the list for the next version.
I have had other request also so I have to do some house keeping before the next version so maybe I can get some input from users as to what can be removed, Hanover is a big chunk.
  • #753
  • May 2, 2012 3:33am May 2, 2012 3:33am
Quoting seller9
Its possible I may continue with the etb EA, but I need to clean out some stuff that is not being used to make room, the Max spread is real easy to add so will put it on the list for the next version.
I have had other request also so I have to do some house keeping before the next version so maybe I can get some input from users as to what can be removed, Hanover is a big chunk.
Hi Seller, that's great news.

I don't use Hannover, I'm trying to keep as simple as possible.
I'll post in more details what I'm using and what I think is dispensable (IMHO).
Thanks

Baffa
  • #754
  • May 2, 2012 9:33am May 2, 2012 9:33am
Quoting seller9
Its possible I may continue with the etb EA, but I need to clean out some stuff that is not being used to make room, the Max spread is real easy to add so will put it on the list for the next version.
I have had other request also so I have to do some house keeping before the next version so maybe I can get some input from users as to what can be removed, Hanover is a big chunk.
Seller,

You might want to consider develping a system with several modules. Main module will contain all the necessary task that is common to all the sub system. All a user need to do is to link main module to the other whichever system they want to trade. jmw1970 in his trade http://www.forexfactory.com/showthread.php?t=313360 is a good example, where a Generator could be linked to each system. Just my suggestion.

Edd
  • #755
  • May 2, 2012 1:52pm May 2, 2012 1:52pm
Quoting seller9
Its possible I may continue with the etb EA, but I need to clean out some stuff that is not being used to make room, the Max spread is real easy to add so will put it on the list for the next version.
I have had other request also so I have to do some house keeping before the next version so maybe I can get some input from users as to what can be removed, Hanover is a big chunk.
i'll give you my input before the weekend, gonna re-examine the settings, it's gonna be easier now since it's all re-organised

Good Job
Don't forget Practice + Patience + Persistence = Profits
  • #756
  • Edited 7:23am May 3, 2012 6:42am | Edited 7:23am
Quoting seller9
Here are some settings posted by Icemaan that you may want to try if your giving this EA a try.

Attachment
Dear Seller9, as you see in the attached examples the time filter seems to be extremely profitable and need to be seriously implemented in the ea.
Have a look to these renko bars charts (4pips boxed) with an ICHi3-3-13 and a time filter of 180''. Imagine a filter based on average time data instead of fixed values!

  1. red dotted vertical lines show bad entries avoided by ea because filtered by time (fixed values: candle time duration>180'')
  2. aqua dotted lines show good entries taken by ea (fixed values: candle time duration<180'')
  3. yellow ones show bad entries not filtered by ea also if filtered by time
  4. purple ones indicate good entries not considered by ea because not well filtered by time


Attached Image (click to enlarge)
Click to Enlarge Name: 2012_05_03_renko b4_ichi3-3-13_usdjpy.gif Size: 29 KB


Attached Image (click to enlarge)
Click to Enlarge Name: 2012_05_03_renko b4_ichi3-3-13_eurusd_correct.gif Size: 33 KB
Ubi maior minor cessat
  • #757
  • May 3, 2012 8:35am May 3, 2012 8:35am
This is an other confirmation of the good improvement given by time filter.
NOTE for Seller9: the ea version 10 is capable to entry but not to fix s/l or BEJS. Why?
Attached Image (click to enlarge)
Click to Enlarge Name: 2012_05_03_renko b4_ichi3-3-13_eurusd_new entry.gif Size: 32 KB
Ubi maior minor cessat
  • #758
  • May 3, 2012 7:48pm May 3, 2012 7:48pm
Quoting Lock
Dear Seller9, as you see in the attached examples the time filter seems to be extremely profitable and need to be seriously implemented in the ea.

Have a look to these renko bars charts (4pips boxed) with an ICHi3-3-13 and a time filter of 180''. Imagine a filter based on average time data instead of fixed values![list][*]red dotted vertical lines show bad entries avoided by ea because filtered by time (fixed values: candle time duration>180'')[*]aqua dotted lines show good entries taken by ea (fixed values: candle time duration<180'')[*]yellow...
I noticed the "RangeBarTime_plus" indicator is different than the one posted, did you modify the one posted?

Also did you add the vertical lines or did the indicator do that and what was the values for the different colored lines? You said aqua was a good signal, does that mean the time was less than 180 for the aqua line?

I am going back through what you posted and see what I can do to add your idea to the EA.

What filter options would you want added?

Thanks
  • #759
  • May 4, 2012 1:04am May 4, 2012 1:04am
Is this something like you were looking for?
The first one just gives the time and the second one is a SMA of the time.

I hope these work for you as they were compiled with Build 419

Attached Image (click to enlarge)
Click to Enlarge Name: chart.jpg Size: 80 KB


Attached File(s)
File Type: ex4 RB_Time.ex4   2 KB | 494 downloads


Attached File(s)
File Type: ex4 RB_sma.ex4   5 KB | 527 downloads
  • #760
  • May 4, 2012 1:09am May 4, 2012 1:09am
Quoting seller9
I noticed the "RangeBarTime_plus" indicator is different than the one posted, did you modify the one posted?

Also did you add the vertical lines or did the indicator do that and what was the values for the different colored lines? You said aqua was a good signal, does that mean the time was less than 180 for the aqua line?

I am going back through what you posted and see what I can do to add your idea to the EA.

What filter options would you want added?

Thanks

Hi Seller9,
I only added the code by Brotyboysome more buffer to have more time periods but finally I didn't use them and then the original code works as normal but with different colors and non standard time settings.
This is my extern setting to filter eurusd renko bars4, ichi 3-3-13:

  1. eintlevel1secs=180 (yellow-gold)
  2. eintlevel2secs=360 (firebrick)

The dotted lines are manually added only to underline the possible filtering action of time. Aqua color as you see correspond to time<180'' and yellow-gold colors of rangebartime.
I'm trying to obtain ($$) a special indicator from the job mql5 community based on a time histogram plus double time moving average to find good momentum working as filter for kumo breakout. (like this) EDIT: you're faster!!!!
But, as a start point and waiting for this ind., we can use fixed time values to filter the ea. It really seems to give the ea a quite reasonable filter action.
An other improvement could come from volumes and we can imagine a special algorithm capable to consider time and volume in an integrated formula. But this could be done in the future.
Have a good day!

Some new bad eurusd positions avoided by time and the last good one on usdjpy

Attached Image (click to enlarge)
Click to Enlarge Name: 2012_05-6_03_renko b4_ichi3-3-13_eurusd_bad positions avoided by time.gif Size: 34 KB


Attached Image (click to enlarge)
Click to Enlarge Name: 2012_05_03_renko b4_ichi3-3-13_usdjpy_new good filter.gif Size: 35 KB
Ubi maior minor cessat
5 traders viewing now