The Dragon | Page 20 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Enter The Dragon - GBP/JPY 2 replies

Info on the Borihei-Dragon Scalping System? 15 replies

Ramadas' Dragon Framework journal 16 replies

Scalping The Dragon 34 replies

***Dragon II*** 15 replies

Attachments: The Dragon
Exit Attachments

The Dragon

  • #381
  • Jun 3, 2010 1:19pm Jun 3, 2010 1:19pm
Quoting jiva34
Taiyakixz,
Greetings..
Noticed the ea has taken some trades against the macd and 10 ma.
I think itmay be the fractals are overriding.
Perhaps the fractals should be set for 1 mi lower fractal (pa going up)
and 5 upper fractal (pa going down) rather using 5 min and 15 min...
Other alternative is to s omehow give preference to the macd and ma over the fractals.
I prefer changing the setting to 1m and 5m fractals...
What say you.
The results were promising.
I did not have gmt set quite right. Entry bar/candle should...
I tried to make the changes in the code but am not sure I did it correctly running the test may not have produced anything reliable.

I just thought maybe it was the fractals causing the trades that were against the ma and macd, but maybe it is something else.
Do you you have any idea why a long would execute when ma and macd are down and vice versa?
If we get this fixed losing trades should be cut way back..
I will try gmt 1 for testing...even as it is and will be eager for the changed codes.

also:
with the code the way it is (5m fractal and 15 min fractal) can we test using 1hr chart like Jeoff's ea? or other time frames or is there something in the code that makes it impractical to test time frames other than 15 min??
  • #382
  • Jun 3, 2010 1:23pm Jun 3, 2010 1:23pm
Quoting Taiyakixz
I will change the code to use 1m and 5m fractals and upload it later.
If you set GMTOffset to 1 hour earlier, EA looks sma and macd at 7:30est.
It makes sense....?
7:30 am est is entry candle 1 hr before data releases maybe too early. Better entry candle 1/2 hr or 1 hr after data release? 9:30, even 10:00? Sometimes trend reverses if data goes against expectations.
  • #383
  • Jun 3, 2010 1:25pm Jun 3, 2010 1:25pm
can you program in a profit target option? I think the trade is simply closing on the trailing stop. Even so if that is the case the results were surprising good. With no target only closing on trailing stop!!!
You are very good programmer I see. Maybe I can learn something...
Thank you.
  • #384
  • Jun 3, 2010 5:26pm Jun 3, 2010 5:26pm
Quoting jiva34
I tried to make the changes in the code but am not sure I did it correctly running the test may not have produced anything reliable.

I just thought maybe it was the fractals causing the trades that were against the ma and macd, but maybe it is something else.
Do you you have any idea why a long would execute when ma and macd are down and vice versa?
If we get this fixed losing trades should be cut way back..
I will try gmt 1 for testing...even as it is and will be eager for the changed codes.

also:
with the code the...
EA looks the direction of ma and macd at 8:30est, then takes order between 9am to 10am.
So if ma and macd are up at 8:30, EA will take long in spite of current ma and macd down.
This is the reason why EA takes the opposite order with your posted chart.

EA is only for 15m. I will change it, but it will take some time. Please wait.
  • #385
  • Jun 3, 2010 6:41pm Jun 3, 2010 6:41pm
EA looks the direction of ma and macd at 8:30est, then takes order between 9am to 10am.
So if ma and macd are up at 8:30, EA will take long in spite of current ma and macd down.
This is the reason why EA takes the opposite order with your posted chart.

8:30 is too early. data comes out at 8:30. takes little while for trend to establish after news. Can we change that? maybe ea looks at 9:00?

is there a way to adjust for that as an option or is it written into the code?

I can wait. I hope this is not too much demand for you. Please do not feel pressured. I know it takes careful thought and much time to write these codes. Whatever you are able to do is nothing short of a miracle.

Jeoff's ea catches Tokyo open. But does not work so well for 8:30 data releases. If we can come up with ea to catch 8:30 like his catches Tokyo that will be something!!

I am putting every spare moment into study writing the codes. Still looks very much like Greek to me but little bit little bit.
  • #386
  • Jun 4, 2010 6:16am Jun 4, 2010 6:16am
jiva34,

EA updated.

- EA looks 5m and 1m fractals.
- add parameter TakeProfit
- add parameter LookBackBars
- still only for 15m chart

With LookBackBars, you can set which bar EA checks the direction.
EA checks ma and macd direction at LookBackBars earlier than starting trade hour.
trade hour starts at 9am, and the default LookBackBars is 1.
So EA checks ma and macd at 8:45 that is 1 bar earlier than 9am.
If you want to use ma and macd at 9am, set to LookBackBars to 0.
If you want to use ma and macd at 8:30, set to LookBackBars to 2.
In case of LookBackBars 0, the direction of ma and macd can change between 9am to 9:15, take care.


To other members,
this EA is for test and under discussing with jiva34. Do not use it with real trade.
Attached File(s)
File Type: mq4 The Dragon EA testv2.mq4   8 KB | 272 downloads
  • #387
  • Jun 4, 2010 1:41pm Jun 4, 2010 1:41pm
Quoting Taiyakixz
jiva34,

EA updated.

- EA looks 5m and 1m fractals.
- add parameter TakeProfit
- add parameter LookBackBars
- still only for 15m chart

With LookBackBars, you can set which bar EA checks the direction.
EA checks ma and macd direction at LookBackBars earlier than starting trade hour.
trade hour starts at 9am, and the default LookBackBars is 1.
So EA checks ma and macd at 8:45 that is 1 bar earlier than 9am.
If you want to use ma and macd at 9am, set to LookBackBars to 0.
If you want to use ma and macd at 8:30, set to LookBackBars to 2.
In case...
you can program like this??????
http://t1.gstatic.com/images?q=tbn:9.../amazement.jpg

I'm on it right away...
  • #388
  • Edited 7:09pm Jun 4, 2010 2:05pm | Edited 7:09pm
these are generally big moves at US data release.
set for bigger profit target and bigger stop loss decent results for EURJPY for 1 yr. want to test other pairs as well.
tsl 80
profit tgt 120
Attached Image (click to enlarge)
Click to Enlarge Name: TesterGraph1.gif Size: 9 KB
  • #389
  • Edited 7:09pm Jun 4, 2010 2:06pm | Edited 7:09pm
1 yr
set for regular stop 80
tsl 80
profit tgt 120

looks like this ea can catch the trend but needs to have wiggle room so as not to get stopped out too early either stopped in loss or stopped in profit

looks like changing the fractals was the right thing...
Attached Image (click to enlarge)
Click to Enlarge Name: TesterGraph2.gif Size: 9 KB
  • #390
  • Edited 7:09pm Jun 4, 2010 2:11pm | Edited 7:09pm
1 yr
just keeps getting better
even bigger profit target
profit target 200
stop 80
tsl 80
Attached Image (click to enlarge)
Click to Enlarge Name: TesterGraph3.gif Size: 9 KB
  • #391
  • Edited 7:08pm Jun 4, 2010 2:14pm | Edited 7:08pm
1 yr
just for the %#$@ of it
stop 80
tsl 80
profit tgt 300
Attached Image (click to enlarge)
Click to Enlarge Name: TesterGraph4.gif Size: 9 KB
  • #392
  • Edited 6:56pm Jun 4, 2010 2:45pm | Edited 6:56pm
macd and 10sma may be even a better combination than 197sma and 10 sma.
may not even be necessary to add robby dss stoch... or heiken ashi..
then again maybe adding robby dss stoch will eliminate the losing trades altogether.

after adding the robby dss to the tested chart the indi did not filter out any losing trades but in fact prevented some winning trades, even after trying several different settings. I think it would confuse things.

Data is released usually at 8:30 am and that can cause a significant move but sometimes there is important data released again at 10:am. So the market can reverse at 10 or just stand still at 8:30 waiting for 10. So there will be some losing trades. How to minimize losses and maximize gains. Big profit target seems to work nicely. Some of these moves can be 200 to 300 pips in a day.
Opening 2 positions and closing one after reasonable gains will help. Like Jeoff's eav9.

want to test this pair EURJPY with dragon algorithm for Jeoff's Tokyo open.

would like to test GBPUSD with dragon algorithm at Tokyo open
can the ea be tested on pairs other than EURJPY?

Great work, Taiyakixz, genius at work.. have a great weekend...
http://t3.gstatic.com/images?q=tbn:k...BJ-preview.jpg
  • #393
  • Jun 4, 2010 3:08pm Jun 4, 2010 3:08pm
Great writing style I love it.

Well done on taming the dragon, he is a mighty beast indeed.

Excellent


Mike
  • #394
  • Jun 4, 2010 3:16pm Jun 4, 2010 3:16pm
Quoting fxsignal
Great writing style I love it.

Well done on taming the dragon, he is a mighty beast indeed.

Excellent


Mike
thanks it's been fun.
And now also profitable with an ea on the way to take all the stress out of it...
Many many thanks to the hard work of Taiyakixz, genius at work!!
  • #395
  • Jun 4, 2010 7:23pm Jun 4, 2010 7:23pm
by the way what causes "order send error #4107?"
How can it be fixed?
  • #396
  • Jun 5, 2010 5:31am Jun 5, 2010 5:31am
Quoting jiva34
thanks it's been fun.
And now also profitable with an ea on the way to take all the stress out of it...
Many many thanks to the hard work of Taiyakixz, genius at work!!
Congratulation jiva34 and Taiyakixz on an excellent thread!

I like the way you throw ideas back and forth to improve the EA.

Keep up the good work!

I gave you 5 stars.
  • #397
  • Jun 5, 2010 11:52am Jun 5, 2010 11:52am
Quoting geoffb65
Congratulation jiva34 and Taiyakixz on an excellent thread!

I like the way you throw ideas back and forth to improve the EA.

Keep up the good work!

I gave you 5 stars.
coming from you, that truly means something!!!
  • #398
  • Jun 7, 2010 5:57am Jun 7, 2010 5:57am
will be testing live demo.
it may be that every day has to be considered differently
check the calendar
set up dragon for 8:30 data when that is the predominant data release
set up dragon latter for 10 data release when that is predominant.
it may be the stops and targets will need to be changed for a 10 data release day.
will be testing for this live demo.
  • #399
  • Jun 7, 2010 6:01am Jun 7, 2010 6:01am
today there is not data release 8:30 or 10
I will not set up today just watch the action.
  • #400
  • Jun 7, 2010 6:10am Jun 7, 2010 6:10am
looks like Thursday and Friday are the days for data this week.
So no setting up until then..
Maybe will play around a bit to see if Dragon ea can catch moves around London open...
the ea is set up for EURJPY at this point and only for 15 min chart.
0 traders viewing now