Gamestudio Links
Zorro Links
Newest Posts
Blobsculptor tools and objects download here
by NeoDumont. 03/28/24 03:01
Issue with Multi-Core WFO Training
by aliswee. 03/24/24 20:20
Why Zorro supports up to 72 cores?
by Edgar_Herrera. 03/23/24 21:41
Zorro Trader GPT
by TipmyPip. 03/06/24 09:27
VSCode instead of SED
by 3run. 03/01/24 19:06
AUM Magazine
Latest Screens
The Bible Game
A psychological thriller game
SHADOW (2014)
DEAD TASTE
Who's Online Now
3 registered members (AndrewAMD, TipmyPip, Edgar_Herrera), 804 guests, and 4 spiders.
Key: Admin, Global Mod, Mod
Newest Members
sakolin, rajesh7827, juergen_wue, NITRO_FOREVER, jack0roses
19043 Registered Users
Previous Thread
Next Thread
Print Thread
Rate Thread
Page 3 of 5 1 2 3 4 5
Re: Luxor system - Jaekle and Tomasini [Re: SFF] #422650
05/14/13 10:27
05/14/13 10:27
Joined: Jul 2000
Posts: 27,977
Frankfurt
jcl Offline

Chief Engineer
jcl  Offline

Chief Engineer

Joined: Jul 2000
Posts: 27,977
Frankfurt
For adapting the time period, you can use the DominantPeriod function or - with the new Zorro - the Spectrum function. As to the questions whether you'll become a millionaire with this new scalping robot, just buy it and find out wink. Robot vendors are poor and need your financial support.

Re: Luxor system - Jaekle and Tomasini [Re: jcl] #422652
05/14/13 10:47
05/14/13 10:47
Joined: Apr 2013
Posts: 107
UK
G
Geek Offline
Member
Geek  Offline
Member
G

Joined: Apr 2013
Posts: 107
UK
I would never consider buying any of these scalping Robots. I just simply do not trust them, i can only trust my own strategies after vigorous back-testing and apart from that creating your own script in Zorro is so much more rewarding, especially when you start getting great results running it live in demo mode etc.

My own methods of tackling the markets using Zorro and understanding how it works is what I am dedicated to... I am still in training and learning mode but I have finally written a scalping short and long script after a month of finding Zorro and the results I am really pleased with..

Looking forward running them live in a few months after some fine tuning and more vigorous back/live testing whilst the other half moans that I am spending to much time playing with my BOT. wink

Re: Luxor system - Jaekle and Tomasini [Re: Geek] #422654
05/14/13 11:19
05/14/13 11:19
Joined: Nov 2012
Posts: 209
S
SFF Offline
Member
SFF  Offline
Member
S

Joined: Nov 2012
Posts: 209
Hi Geek,

What results did you get your own bot?
I am very interested.

Re: Luxor system - Jaekle and Tomasini [Re: Geek] #422655
05/14/13 11:23
05/14/13 11:23

A
acidburn
Unregistered
acidburn
Unregistered
A



Originally Posted By: Geek
I would never consider buying any of these scalping Robots. I just simply do not trust them, i can only trust my own strategies after vigorous back-testing and apart from that creating your own script in Zorro is so much more rewarding, especially when you start getting great results running it live in demo mode etc.


Couldn't agree more!

Re: Luxor system - Jaekle and Tomasini [Re: SFF] #422658
05/14/13 13:08
05/14/13 13:08
Joined: Apr 2013
Posts: 107
UK
G
Geek Offline
Member
Geek  Offline
Member
G

Joined: Apr 2013
Posts: 107
UK
Originally Posted By: SFF
Hi Geek,

What results did you get your own bot?
I am very interested.


Well, currently I have been backtesting each year individually and doing some walk forward testing and oversampling etc. I am running it live on demo now to see how it performs in live trading scenario as the backtesting results are just way to good to be true... Last night i put the finishing touches to it which i am happy with, and it works great on most currency pairs 500%-3000%+ and even gold on the backtesting and out of sample / walk forward testing etc.

The aim was to write a script to return me $100 per day average with minimal draw down, minimal capital, decent winning trades, high profit, not so much risk etc.

This is the the performance report I have achieved from my just my shorting script alone, as you can see, the annual return, profit factor, draw-down etc is all pretty amazing, the log file data seems fine as does everything else.

This has not yet been optimized at all either: I am looking forward to seeing how the live demo works and if it does work as good as it has backested. I really hope so as currently I am a Zorro believer and optimistic in my own methods of tackling these markets. Let's see how it goes, but IF it does work as good as the backtest and these kind of returns are indeed possible for the average person with limited programming knowledge who is not a hedge fund or financial institution making billions then i will be a very happy man.

Quote:
Walk-Forward Test MYBOT11 AUD/USD - performance report

Simulation period 03.01.2007-31.12.2012
Test period 25.09.2008-31.12.2012
WFO test cycles 14 x 7543 bars (16 weeks)
Training cycles 15 x 42743 bars (92 weeks)
Lookback time 150 bars (37 hours)
Assumed spread 2.5 pips (roll 0.41/-0.86)
Assumed slippage 10.0 sec
Contracts per lot 800

Gross win/loss 171108$ / -61444$ (+1370796p)
Average profit 25702$/year, 2142$/month, 99$/day
Max drawdown -835$ 1% (MAE -827$ 1%)
Total down time 51% (TAE 11%)
Max down time 15 days from Apr 2011
Largest margin 400$
Trade volume 232972274$ (54601564$/year)
Transaction costs -60988$ spr, -407$ slp, -513$ rol
Capital required 1101$

Number of trades 7624 (1786/year)
Percent winning 71%
Max win/loss 513$ / -553$
Avg trade profit 14$ 179.8p (+32$ / -27$)
Avg trade bars 2 (+2 / -3)
Max trade bars 13 (3 hours)
Time in market 22%
Max open trades 2
Max loss streak 9 (uncorrelated 7)

Annual return 3669%
Profit factor 2.78 (PRR 2.69)
Sharpe ratio 11.19
Kelly criterion 5.43
OptimalF .007
Ulcer index 0%
Prediction error 17%
Cycle performance 2.88 2.76 2.82 2.67

Portfolio analysis OptF ProF Win/Loss Cycles

AUD/USD:L .000 ---- 0/0 ..............
AUD/USD:S .007 2.67 5239/2193 //////////////




Last edited by Geek; 05/14/13 13:14.
Re: Luxor system - Jaekle and Tomasini [Re: Geek] #422659
05/14/13 13:22
05/14/13 13:22

A
acidburn
Unregistered
acidburn
Unregistered
A



Your results definitely look too good to be true. Are you sure you haven't accidentaly leaked a bit of future knowledge in your algorithm? If your results are genuine, you are going to be infinitely rich in a very short time. cool

This reminds me of the times when I was more actively playing with renko bars. At one point I got so good strategy backtest results that I almost fell of the chair. grin Only a few hours later I noticed that I accidentaly introduced a very small lookahead window while _generating_ the bricks. The bricks itself contained a bit of knowledge of the future. No wonder almost any strategy I run on that data set was being insanely profitable. blush

Re: Luxor system - Jaekle and Tomasini [Re: ] #422660
05/14/13 13:34
05/14/13 13:34
Joined: Apr 2013
Posts: 107
UK
G
Geek Offline
Member
Geek  Offline
Member
G

Joined: Apr 2013
Posts: 107
UK
Thanks, I just took away the LookBack, and it returns the same, and can't see anything in the code to suggest it has the knowledge of the future.

I've been working solid on it so I hope this works in live demo testing.

A few weeks ago I had a 100,000% increase on a backtest and nearly did the same as you but then when running it live it didn't work and it was attributed to the future knowledge as you describe.

Last edited by Geek; 05/14/13 13:39.
Re: Luxor system - Jaekle and Tomasini [Re: ] #422661
05/14/13 13:37
05/14/13 13:37
Joined: Jul 2000
Posts: 27,977
Frankfurt
jcl Offline

Chief Engineer
jcl  Offline

Chief Engineer

Joined: Jul 2000
Posts: 27,977
Frankfurt
I have to agree to acidburn, the result looks wrong. Even worse, it looks like a bug in Zorro, because accidental future peeking is normally prevented by the program structure unless you've set the PEEK flag or changed some parameters between training and testing.

Check the result in detail before trading it live. Examine some single trades from entry to exit. If you find something that is not right, let me know and I'll look into it.

Re: Luxor system - Jaekle and Tomasini [Re: jcl] #422663
05/14/13 13:42
05/14/13 13:42
Joined: Apr 2013
Posts: 107
UK
G
Geek Offline
Member
Geek  Offline
Member
G

Joined: Apr 2013
Posts: 107
UK
I hope not jcl, i will keep this running in demo and go look over everything in detail...

It seems to be working fine at the moment in live demo mode, but i see will how how it exits the trades etc.

Thanks.

Re: Luxor system - Jaekle and Tomasini [Re: Geek] #422761
05/16/13 10:07
05/16/13 10:07
Joined: Nov 2012
Posts: 209
S
SFF Offline
Member
SFF  Offline
Member
S

Joined: Nov 2012
Posts: 209
Hi Geek,

I would like to know how money management improves your bot.
I think that the result you posted above is without MM.

Last edited by SFF; 05/16/13 10:07.
Page 3 of 5 1 2 3 4 5

Moderated by  Petra 

Gamestudio download | chip programmers | Zorro platform | shop | Data Protection Policy

oP group Germany GmbH | Birkenstr. 25-27 | 63549 Ronneburg / Germany | info (at) opgroup.de

Powered by UBB.threads™ PHP Forum Software 7.7.1