Attachments: Trading FRAMA Ribbon only

Trading FRAMA Ribbon only

#21
Oct 12, 2019 9:53am Oct 12, 2019 9:53am

northwind
Joined Oct 2017 | Status: Trader | 362 Posts

Quoting northwind

Disliked

Hi Joe, Woke up this morning and found a pleasant surprise. Thank you for the fast implementation. Really impressive. A few days ago I ran script ForceLoadHistoricalData. ..... Can anyone show where I went wrong in Strategy Tester? {image} {image} {image} {image} {image}

Ignored

Found the problem. Tester data was old.
Reran ForceLoadHistricalData.
When run ForceLoadHistricalData the new data gets saved in an incremented new file. I doesn't replace previous data file.
For example:
EURUSD60_0.fxt is the data tester uses,
but EURUSD60_2.fxt is the most recent ForceLoadHistoricalData data.

Perfidious Pips

#22
Oct 12, 2019 10:02am Oct 12, 2019 10:02am

JoeeF
Joined Aug 2010 | Status: Life is easy, life is delightful. | 3,152 Posts

Quoting northwind

Disliked

{quote} Found the problem. Tester data was old. Reran ForceLoadHistricalData. When run ForceLoadHistricalData the new data gets saved in an incremented new file. I doesn't replace previous data file. For example: EURUSD60_0.fxt is the data tester uses, but EURUSD60_2.fxt is the most recent ForceLoadHistoricalData data.

Ignored

And you will leave pain to take care of itself.

#23
Oct 12, 2019 10:40am Oct 12, 2019 10:40am

JoeeF
Joined Aug 2010 | Status: Life is easy, life is delightful. | 3,152 Posts

Quoting JoeeF

Disliked

{quote} {quote} Installation: 1) Put indicator (frama ribbon 1.1 (arrows).ex4) in indicators folder: \MQL4\Indicators 2) Put EA (JFFramaRibbon.ex4) in experts folder: \MQL4\Experts 3) Put XML config (JFFramaRibbon.xml) in common files folder: \MetaQuotes\Terminal\Common\Files (If someone isn't able to locate common files folder please look here.) That's it. Notes: 1) MT5...

Ignored

I've omitted inputs in previous version. So, a new version is here. It's only a little improvement for better experience in strategy tester.
Only replace EA with a new one.

Attached Image (click to enlarge)

Click to Enlarge

Name: Image2.png
Size: 27 KB

Attached Image

Attached File(s)

JFFramaRibbon_v1.01_MT4.rar 254 KB | 549 downloads

JFFramaRibbon_v1.01_MT4.zip 253 KB | 511 downloads

And you will leave pain to take care of itself.

#24
Oct 12, 2019 1:13pm Oct 12, 2019 1:13pm

northwind
Joined Oct 2017 | Status: Trader | 362 Posts

Thanks Joe,
Ran some tests with new version 1.01.
The new interface is great.
Next task is to get tick data for testing.

Attached Image(s) (click to enlarge)

Click to Enlarge

Name: Image1.png
Size: 16 KB

Click to Enlarge

Name: Image3.png
Size: 17 KB

Click to Enlarge

Name: Image4.png
Size: 30 KB

Attached Image

Perfidious Pips

#25
Oct 12, 2019 1:19pm Oct 12, 2019 1:19pm

northwind
Joined Oct 2017 | Status: Trader | 362 Posts

Attached Image (click to enlarge)

Click to Enlarge

Name: Image1.png
Size: 17 KB

Attached Image

Perfidious Pips

#26
Oct 12, 2019 2:15pm Oct 12, 2019 2:15pm

northwind
Joined Oct 2017 | Status: Trader | 362 Posts

Still need to get tick data.

Ran strategy tester with optimizer on TP & SL.
Configured with optimized TP & SL for frama period 13:

Attached Image

Results:

Attached Image (click to enlarge)

Click to Enlarge

Name: Image2.png
Size: 30 KB

Perfidious Pips

#27
Oct 12, 2019 4:23pm Oct 12, 2019 4:23pm

northwind
Joined Oct 2017 | Status: Trader | 362 Posts

After getting optimized frama period = 14, ran again to optimize TP & SL.
Results are SL = 160 and TP ' 110.
Strategy Tester with frama period = 14, TP = 110 and SL = 160.

Attached Image(s) (click to enlarge)

Click to Enlarge

Name: Image1.png
Size: 30 KB

Click to Enlarge

Name: Image2.png
Size: 18 KB

Perfidious Pips

#28
Oct 13, 2019 12:00am Oct 13, 2019 12:00am

northwind
Joined Oct 2017 | Status: Trader | 362 Posts

GBPUSD H1 Optimizer results:

Attached Image(s) (click to enlarge)

Click to Enlarge

Name: Image2.png
Size: 13 KB

Perfidious Pips

#29
Oct 13, 2019 9:03am Oct 13, 2019 9:03am

northwind
Joined Oct 2017 | Status: Trader | 362 Posts

Hi Joe,

Great job. Thank you for your EA.
The new user interface makes it much, much easier and faster to use in Strategy Tester.
It runs without a hitch in Strategy tester and matches buys & sells on a regular chart.
To improve winning profit percent, can the EA be programmed to incrementally add to winning trades?
For example add-on BUY:
When EA takes a BUY and price moves x-points up, add one additional BUY, then price moves x-points up, add one additional BUY, etc.
User sets the value for x-points and lot size for incremental BUYS.
All BUYS have the same TP and SL as the original BUY.

For example add-on SELL:
When EA takes a SELL and price moves x-points down, add one additional SELL, then price moves x-points down, add one additional SELL, etc.
User sets the value for x-points and lot size for incremental SELLS.
All SELLS have the same TP and SL as the original SELL.

Thanks again for your creativity and generous work.
All the best,
John

Perfidious Pips

#30
Oct 13, 2019 10:29am Oct 13, 2019 10:29am

JoeeF
Joined Aug 2010 | Status: Life is easy, life is delightful. | 3,152 Posts

Quoting northwind

Disliked

The new user interface makes it much, much easier and faster to use in Strategy Tester. It runs without a hitch in Strategy tester and matches buys & sells on a regular chart.

Ignored

This is important message.

Quoting northwind

Disliked

To improve winning profit percent, can the EA be programmed to incrementally add to winning trades? For example add-on BUY: When EA takes a BUY and price moves x-points up, add one additional BUY, then price moves x-points up, add one additional BUY, etc. User sets the value for x-points and lot size for incremental BUYS. All BUYS have the same TP and SL as the original BUY. For example add-on SELL: When EA takes a SELL and price moves x-points down, add one additional SELL, then price moves x-points down, add one additional SELL, etc. User sets...

Ignored

I'll look at this.

And you will leave pain to take care of itself.

#31
Oct 13, 2019 10:55am Oct 13, 2019 10:55am

northwind
Joined Oct 2017 | Status: Trader | 362 Posts

Quoting JoeeF

Disliked

{quote} This is important message. {quote} I'll look at this.

Ignored

Hi Joe,
You have done some fantastic work on this EA.
The new user interface is much easier to understand and use than editing XML files.

Perfidious Pips

#32
Oct 13, 2019 11:12am Oct 13, 2019 11:12am

JoeeF
Joined Aug 2010 | Status: Life is easy, life is delightful. | 3,152 Posts

Quoting northwind

Disliked

{quote} Hi Joe, You have done some fantastic work on this EA. The new user interface is much easier to understand and use than editing XML files.

Ignored

Yes. It's much easier to use this EA in strategy tester with this option. MT4 strategy tester has limitation for only one symbol.
However, if someone will be using for example 15 symbols in one EA and more strategies - editing XML will be necessity.

And you will leave pain to take care of itself.

#33
Oct 13, 2019 2:36pm Oct 13, 2019 2:36pm

alblive
| Joined Apr 2019 | Status: Trader | 85 Posts

Quoting northwind

Disliked

Hi Joe, Great job. Thank you for your EA. The new user interface makes it much, much easier and faster to use in Strategy Tester. It runs without a hitch in Strategy tester and matches buys & sells on a regular chart. To improve winning profit percent, can the EA be programmed to incrementally add to winning trades? For example add-on BUY: When EA takes a BUY and price moves x-points up, add one additional BUY, then price moves x-points up, add one additional BUY, etc. User sets the value for x-points and lot size for incremental BUYS. All BUYS...

Ignored

have you considered the EA choosing a lot size based on user defined risk %?

for example, the user wants to risk 1% per trade. the ea would calculate this based on stop loss and account equity.

#34
Oct 13, 2019 2:46pm Oct 13, 2019 2:46pm

northwind
Joined Oct 2017 | Status: Trader | 362 Posts

Quoting alblive

Disliked

{quote} have you considered the EA choosing a lot size based on user defined risk %? for example, the user wants to risk 1% per trade. the ea would calculate this based on stop loss and account equity.

Ignored

Hi alblive,
Thanks for your comments.
A good idea, something to consider for future implementation.
Have to consider Joe's time and effort, which he has generously shared.
At this point I'd like to take it step-by-step and make sure the EA is working.
Please make a note and check back.
Thanks.

Perfidious Pips

#35
Oct 13, 2019 3:19pm Oct 13, 2019 3:19pm

cfudge
Joined Apr 2013 | Status: Trader | 1,570 Posts

Hi John
This is what I meant in the PM

Chris

Attached Image (click to enlarge)

Click to Enlarge

Name: for John.jpg
Size: 128 KB

#36
Oct 13, 2019 3:34pm Oct 13, 2019 3:34pm

JoeeF
Joined Aug 2010 | Status: Life is easy, life is delightful. | 3,152 Posts

Quoting alblive

Disliked

{quote} have you considered the EA choosing a lot size based on user defined risk %? for example, the user wants to risk 1% per trade. the ea would calculate this based on stop loss and account equity.

Ignored

Yes, you're right. You can see empty <STRATEGY_MONEY_MANAGEMENT> tag in config XML. I haven't implemented it till now.
It isn't so easy. I need robust solution for both MT4 and MT5 terminals.
regards

Attached Image (click to enlarge)

Click to Enlarge

Name: Image1.png
Size: 38 KB

And you will leave pain to take care of itself.

#37
Oct 13, 2019 4:35pm Oct 13, 2019 4:35pm

northwind
Joined Oct 2017 | Status: Trader | 362 Posts

Quoting cfudge

Disliked

Hi John This is what I meant in the PM Chris {image}

Ignored

Hi Chris,
Joe has done a terrific job with this EA. All credit goes to him.
Only tested on H1. I guess H4 should be next
I ran a couple different pairs through strategy tester optimizer.
Round 1 was optimize max profit on frama period and the two multipliers. The frama range to optimize over was quite large.
Round 2, using best values from round 1, optimize max profit on TP & SL.
My data isn't the best and only tested/optimized on 5 months of data.
As you know this process takes a considerable amount of time.
Also, after getting a round 1 optimal value for frama range I reran optimizer using micro steps around the first optimum value to try and zero in on the best value. Optimizer didn't find any improvement.
An interesting finding is the larger TP & SL values.
They also arrived at local peak values, getting worse profit when changed higher or lower.
All this is probably an artifact of my crappy data.
Another finding is using TYPICAL Price which eliminates many of the small, quick reversals.

Hey, give it a try and let me know your results, or if you find the holy grail.

Perfidious Pips