DAX 15Min - False Breakout / SuperTrend

NoName

2 years ago

# Reply

Thank you very much for this fascinating trading system. It is still proving to be extremely effective in 2024. It illustrates once again how valuable individual trading systems can be.

Plop61

10 years ago

# Reply

Hello,
Thank you for sharing this beautiful strategy.
Is it possible to indicate the code for short positions?
Best Regard.

danhei

10 years ago

# Reply

Hi

I am tryong to figure out how trist strategy works. I am new to pro real time.

Can someone explain the entry and exit method? Many thanks

Elsborgtrading

10 years ago

# Reply

Hi Triss. the File is in the post for download. There is nothing wrong with the code. Your error describes 2 line- but it empty- perhaps some character sneaked in yours? . Please check that you code is correct copied. or download the file dax-15m-longonly.itf

Cheers Kasper

triss1965@gmail.com

10 years ago

# Reply

why do I get error ond line whit DAX? Eihter cod works. Can you chare the file?

jonjon

10 years ago

# Reply

Hi Kenneth. I can't attach the screenshot I'm afraid. Unless I'm doing something stupid. I pasted the screenshot into Word and tried to attach the file.

Kenneth Kvistad

10 years ago

# Reply

Jonjon, do you have screenshoot of this strategy in tick by tick mode?

jonjon

10 years ago

# Reply

Thanks Maxime. Appreciate it.

I'll let you know how it goes. I've been running it in demo in the background whilst I have been focusing on other strategies. No trades yet though so I'll take another look at the code probably next week.

Cheers

DESCOUMA

10 years ago

# Reply

Hi,

Thanks for your comments and sorry for my late answer :)

m, d & p are optimized variables which I chose to maximize the Profit/Drawdown Ratio. Hence, since September 2012 using my last settings (0.5 %Loss / 1.5 %Profit / No reinvestment / No multipliers) it gives a worst loss of 1.77%, a DD of 12.8% and a profit of 94.17% with gain/loss = 1.51 and 33.21 winning trades.

As all except 1 of the winning trades came after more than 5 candles in this last backtest, my guess is that tick mode or not it won't change many things.

Maxime

jonjon

10 years ago

# Reply

Yes. Well I think I am. I'm running v. 10.3 and I ticked "ProBacktest in tick by tick mode"

K12AN

10 years ago

# Reply

Have you backtested using tick data?

jonjon

10 years ago

# Reply

 This is very cool. Thank you very much.

I've been playing a bit with it. How did you get your m, p and d values? m and p relate to the Supertrend indicator....as I'm new to this indicator can you tell me if these are pretty standard variables? d = 20 seems reasonable to me.

I have just started paper trading the strategy. Adjusted the variables slightly to make a little more conservative: Rg = 0.4, Risk = 0.01, No reinvest. I've also optimised the p and d to 26 and 21 respectively (hence my question above).

Back tested in tick mode from Sept 2012 which is just 60,000 units back (round number, nothing scientific there however thought it appropriate to base test from period post Draghi saying that ECB would do anything to stop Euro failing in July 2012 and then subsequent quieting down of volatility / market).

I can't attach my results (is this as I'm a Junior Member?)

% winning trades = 40%

Gain / Loss: 2.01

I still want to get the loss of the worse trade down (-3,752) and the size of the Maxdrawdown is too large (-19,291 with 8 consecutive losses).

I'll look further into it however anyone looked at the poor performance in Q2 2013 and last quarter of this year? Down periods are to be expected but just wondering whether there was anything macro going on that we should bear in mind going forward.

I'll let you know how my trades go with this. Looking forward to hearing from any other of you guys paper trading this.

Cheers

DESCOUMA

10 years ago

# Reply

Yes. I'll probably give it a try with 1% $risk / no reinvestment until I find a way to reduce the drawdown.

Cosmic1

10 years ago

# Reply

Great work in trying to avoid curve fitting and walk forward test. Good to know that you have done this as gives some confidence moving forward to demo/live test. Are you planning running this on demo, live with minimum stakes or the code posted? 

DESCOUMA

10 years ago

# Reply

Hi!

Thanks for your comments and thanks Reiner for your syntax on seasonality.

I did some walk forward testing and found quite good results particularly in the recent period (see attached .xlsx file).

In my testing, I just managed to solve for the tp and sl (changed from pips to percentages in this study) which give me the best Return/Drawdown ratio using a 2 years period. Then I used those parameters for the next 6 months. And again, and again !

Here is my last version of the code used for this test:

DEFPARAM CumulateOrders = false
//DAX - 15M

ONCE p=18
ONCE d=20
ONCE m=1.5
ONCE Rg = 0.5
ONCE Risk = 0.015
ONCE Monday =2
ONCE Tuesday = 2
ONCE Wednesday = 1
Once Thursday = 1
Once Friday = 1

If Opendayofweek = 1 then
DayMult = Monday
ElsIf Opendayofweek = 2 then
DayMult = Tuesday
ElsIf Opendayofweek = 3 then
DayMult = Wednesday
ElsIf Opendayofweek = 4 then
DayMult = Thursday
ElsIf Opendayofweek = 5 then
DayMult = Friday
Endif

aMax = Highest[d](High)
aMin = Lowest[d](Low)
SupTnd = SuperTrend[m,p]
RngOk = (High-Low)/(aMax-aMin)<Rg
sl=0.5
tp=1.5
Reinv = 1
Coef = 0
Balance = 23827.1
If Reinv = 1 Then
Equity = Balance+StrategyProfit
Else
Equity = 10000
EndIf
If Coef<>1 Then
DayMult=1
EndIf

n = Max(1,Equity*Risk/(Close*Sl/100)*SQRT(DayMult))

cBuy = aMin[1]<aMin[2] And High<aMax[1] And Close>Open And RngOk And Close>SupTnd

If cBuy Then
Buy n Shares at Market
EndIf

SET STOP %loss sl
SET TARGET %PROFIT tp

Maxime

ALE

10 years ago

# Reply

Thank you for your new ideas, it' very good.. !! seasonalty is reiner's idea..but thank you for your Credit!!

GraHal

10 years ago

# Reply

Good Morning Descouma

Thank you so much for your code.

I will get back on here if I come up with any improvements.

Cheers

GraHal