Discussing the strategy VECTORIAL DAX (M5)

Viewing 15 posts - 736 through 750 (of 1,264 total)

← 1 2 3 … 49 50 51 … 83 84 85 →

Author

Posts
05/06/2020 at 7:33 AM #130164 quote

zilliq
Participant

Master

bertrandpinoy wrote:

demo test since April 22. no loss on CAC40.

On 3 trades only Bertrand 😉 But it’s a beginning …

bertrandpinoy thanked this post

05/08/2020 at 10:47 AM #130588 quote
zilliq
Participant

Master
Hi Guys,

Don’t know if someone mentioned it but there is an error in the code

For me it’s not
```
//VECTEUR = CALCUL DE LA PENTE ET SA MOYENNE MOBILE
MMB = Exponentialaverage[PeriodeB](close)
pente = (MMB-MMB[nbchandelierB]*pipsize) / nbchandelierB
trigger = Exponentialaverage[PeriodeB+lag](pente)
```
But
```
MMA = exponentialaverage[PeriodeA](customclose)

ADJASUROPPO = ((MMA-MMA[nbchandelierA])/pipsize) / nbChandelierA
ANGLE = (ATAN(ADJASUROPPO))

trigger = exponentialaverage[PeriodeA+lag](ANGLE)
```
Because in the first code the code is always positive and not positive/negative depends on the angle

Have a nice day

Zilliq
05/09/2020 at 7:29 PM #130794 quote

Paul
Participant

Master

hi zilliq

can you post some results or code using your change?

Thanks

05/09/2020 at 8:24 PM #130804 quote
nonetheless
Participant

Master
I had assumed he meant only to redefine ‘trigger’
```
//STRATEGIE
//VECTEUR = CALCUL DE L'ANGLE
ONCE PeriodeA = 10
ONCE nbChandelierA= 14
MMA = Exponentialaverage[PeriodeA](customclose)
ADJASUROPPO = (MMA-MMA[nbchandelierA]*pipsize) / nbChandelierA
ANGLE = (ATAN(ADJASUROPPO)) //FONCTION ARC TANGENTE

CondBuy1 = ANGLE >= 45
CondSell1 = ANGLE <= - 34


//VECTEUR = CALCUL DE LA PENTE ET SA MOYENNE MOBILE
ONCE PeriodeB = 22
ONCE nbChandelierB= 34
lag = 4
MMB = Exponentialaverage[PeriodeB](close)
pente = (MMB-MMB[nbchandelierB]*pipsize) / nbchandelierB

trigger = exponentialaverage[PeriodeA+lag](ANGLE)
```
but the results are dire. Something else then?
05/10/2020 at 9:20 AM #130837 quote

zilliq
Participant

Master

Paul wrote:

hi zilliq

can you post some results or code using your change?

Thanks

Of course, let me some minutes..

05/10/2020 at 9:21 AM #130838 quote
zilliq
Participant

Master
First, the code in the actual vectorial dax code was
```
//VECTEUR = CALCUL DE LA PENTE ET SA MOYENNE MOBILE
ONCE PeriodeB = 22
ONCE nbChandelierB= 34
lag = 4
MMB = Exponentialaverage[PeriodeB](close)
pente = (MMB-MMB[nbchandelierB]*pipsize) / nbchandelierB
trigger = Exponentialaverage[PeriodeB+lag](pente)
// modifier par fifi743
// ajout d'un filtre pour achat
filtre,weekpivot,pivot,filt =call"FILTRE_Prise_Position"(close)
//===================== ajouter par fifi743
donchianP = 20
hh=highest[donchianP](high)
ll=lowest[donchianP](low)
MA=Average[4](close)
Myrsi=RSI[2](close)
CondBuy2 = (pente > trigger) AND (pente < 0) and high>MA  and coeff>0.9
CondSell2 = (pente CROSSES UNDER trigger) AND (pente > -1)
```
05/10/2020 at 9:25 AM #130839 quote
zilliq
Participant

Master
If you use the actual vectorial dax code you can NEVER have a negative pente and so you can NEVER have a CondBuy2 and the pente is ALWAYS>-1
```
PeriodeB = 22
nbChandelierB= 34
lag = 4
MMB = Exponentialaverage[PeriodeB](close)
pente = (MMB-MMB[nbchandelierB]*pipsize) / nbchandelierB

Return pente
```
2020-05-10_10h24_12.jpg
05/10/2020 at 9:28 AM #130842 quote
zilliq
Participant

Master
And now if you correct the code, you have positive/negative pente and Good Condbuy
```
PeriodeB = 22
nbChandelierB= 34
lag = 4
MMB = Exponentialaverage[PeriodeB](close)
pente = ((MMB-MMB[nbchandelierB])/pipsize) / nbchandelierB

Return pente
```
2020-05-10_10h27_13.jpg
05/10/2020 at 10:28 AM #130857 quote

Francesco
Participant

Veteran

So the correction it’s just the “/” in the pente calculation?

05/10/2020 at 11:33 AM #130864 quote
Paul
Participant

Master
Thanks zilliq for the explaining.

I tested your conclusion
```
pente is ALWAYS>-1
```
so I checked with this
```
if pente<=-1 then
test=1
else
test=0
endif

graph test
```
and got the result in the pic

I’am taking another look with your modification.

Screenshot-2020-05-10-at-12.25.12.jpg
05/10/2020 at 11:47 AM #130868 quote

Paul
Participant

Master

test was ok

05/10/2020 at 11:48 AM #130870 quote

zilliq
Participant

Master

Hi Paul, use the actual code not the correction and as you see on my picture (the blue) the pente is always positive and it’s normal because of the code

05/10/2020 at 11:51 AM #130872 quote
zilliq
Participant

Master
With the actual “mistake” code
```
PeriodeB = 22
nbChandelierB= 34
lag = 4
MMB = Exponentialaverage[PeriodeB](close)
pente = (MMB-MMB[nbchandelierB]*pipsize) / nbchandelierB
 
if pente<=-1 then
test=1
else
test=0
endif
 
return test
```
Always 0 = mistake

2020-05-10_12h50_44.jpg
05/10/2020 at 3:18 PM #130939 quote

Paul
Participant

Master

I cannot reach the same conclusion, even as indicator on the dax 5min timeframe. Maybe have a look at my latest version. I will release it in the next post.

05/10/2020 at 3:51 PM #130951 quote
zilliq
Participant

Master
HI @paul

Take the actual code
```
PeriodeB = 22
nbChandelierB= 34
lag = 4
MMB = Exponentialaverage[PeriodeB](close)
pente = (MMB-MMB[nbchandelierB]*pipsize) / nbchandelierB
```
And you will see you can’t have negative value as pipsize is on MMB[nbchandelierB] and not (MMB-MMB[nbchandelierB])

See U

Paul thanked this post
Author

Posts