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
2 registered members (Quad, aliswee), 835 guests, and 5 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 1 of 2 1 2
MaxDD in % #477622
07/17/19 14:07
07/17/19 14:07
Joined: Jul 2017
Posts: 783
Z
Zheka Offline OP
User
Zheka  Offline OP
User
Z

Joined: Jul 2017
Posts: 783
JCL,

The manual says that this number - in % - is calculated relative to GROSS profit, and not balance or equity - which is the commonly accepted way of doing it.

This of course makes it a much smaller number.

What's the meaning of such a metric?


Also, the numbers exported to a _pnl.csv reflect accumulated profit, without taking into account the starting Capital.
Probably, ok if the Capital is not set, but is not "....the daily equity or balance curve" as the manual suggests.

Re: MaxDD in % [Re: Zheka] #477623
07/17/19 14:14
07/17/19 14:14
Joined: Jul 2000
Posts: 27,977
Frankfurt
jcl Offline

Chief Engineer
jcl  Offline

Chief Engineer

Joined: Jul 2000
Posts: 27,977
Frankfurt
If performance figures were relative to balance or equity, a rich guy with much money on his account would get a different performance than a poor guy. wink You normally want to know the performance of your strategy, not of your account.

The exception is testing the money management. Then you really need metrics relative to equity, such as the CAGR. For this set up the "Capital" variable. All metrics change then their meaning. Read the "Performance" chapter - all is explained there.

Re: MaxDD in % [Re: Zheka] #477624
07/17/19 14:51
07/17/19 14:51
Joined: Jul 2017
Posts: 783
Z
Zheka Offline OP
User
Zheka  Offline OP
User
Z

Joined: Jul 2017
Posts: 783

I am testing the money management and the "Capital" variable is set. How is MaxDD in % is then calculated?

I cannot see anything describing difference in calculations of this number in the "Performance chapter" - when Capital is set and when not.

Right now, it looks like % is being calculated relative to Gross Wins, rather than Gross Profit.
It is appr 2x less than the number calculated manually from a _pnl.csv (with addition of Capital), even with the BALANCE flag set.


As for numbers in _pnl.csv: shouldn't they start from the initial Capital when Capital var is set?

Re: MaxDD in % [Re: Zheka] #477626
07/17/19 15:49
07/17/19 15:49
Joined: Jul 2000
Posts: 27,977
Frankfurt
jcl Offline

Chief Engineer
jcl  Offline

Chief Engineer

Joined: Jul 2000
Posts: 27,977
Frankfurt
Just look in the manual: "Max DD - Maximum drawdown during the test period, in account currency units and in percent from the gross profit.".

Reinvesting capital will of course affect the gross profit, which renders the max DD relatively meaningless. It depends then not anymore on the strategy performance, but on your reinvesting. For knowing the drawdown of your strategy, disable reinvesting.

"PnL" means "Profit and Loss", not "Profit plus Capital".

Re: MaxDD in % [Re: Zheka] #477630
07/17/19 16:28
07/17/19 16:28
Joined: Jul 2017
Posts: 783
Z
Zheka Offline OP
User
Zheka  Offline OP
User
Z

Joined: Jul 2017
Posts: 783
Sure, maxDD in absolute in meaningless for a reinvesting strategy. But I am talking about a percentage calculation!

How is MaxDD calculated as a percentage when Capital var is set?

Re: MaxDD in % [Re: Zheka] #477637
07/18/19 08:43
07/18/19 08:43
Joined: Jul 2000
Posts: 27,977
Frankfurt
jcl Offline

Chief Engineer
jcl  Offline

Chief Engineer

Joined: Jul 2000
Posts: 27,977
Frankfurt
By dividing it by the gross profit. - It is arguable if a DD percentage has any meaning with Capital set, but if so, then for the reinvesting, not for the strategy. You can get any DD percentage dependent on how you reinvest. So better check max DD and the other figures first without Capital. This reflects the real performance of your strategy.

Re: MaxDD in % [Re: jcl] #477638
07/18/19 09:17
07/18/19 09:17
Joined: Jul 2017
Posts: 783
Z
Zheka Offline OP
User
Zheka  Offline OP
User
Z

Joined: Jul 2017
Posts: 783
Originally Posted by jcl
It would be probably better to use the preceding profit peak at the time of the drawdown instead
Of course, this is its definition ( https://www.investopedia.com/terms/m/maximum-drawdown-mdd.asp)
Dividing by the total gp (at the end of a simulation) just has no meaning, in any case.
Without re-investment, maxDD reporting in % - IMHO - should be omitted altogether.

Re: MaxDD in % [Re: Zheka] #477639
07/18/19 09:45
07/18/19 09:45
Joined: Jul 2017
Posts: 783
Z
Zheka Offline OP
User
Zheka  Offline OP
User
Z

Joined: Jul 2017
Posts: 783
Also:
- when plotting results, it would make sense to display percentage DD for the case with reinvestment, and absolute DD numbers - without.
- report 'closed trade' (or balance-to-balance) DD.

- report average (or median) DD above some min threshold (of %, $ or duration)
- report average of top 5 DDs

These will be more informative than just one 1 maxDD number, and too burdensome for users to calculate themselves.

Last edited by Zheka; 07/18/19 09:56.
Re: MaxDD in % [Re: Zheka] #477640
07/18/19 11:19
07/18/19 11:19
Joined: Jul 2000
Posts: 27,977
Frankfurt
jcl Offline

Chief Engineer
jcl  Offline

Chief Engineer

Joined: Jul 2000
Posts: 27,977
Frankfurt
I had posted the idea with the preceding peak, but then realized that it's a bad idea. The same max drawdown would then get a different percentage dependent on when in the simulation it happend. So, this won't work.

I understand that there are 100 different performance metrics, but we want to concentrate on the useful and relevant ones. You can easily calculate your favorite DD variant in the simulation and then print them to the performance report. Further suggestions are welcome - please post them to the Zorro future forum.

Re: MaxDD in % [Re: jcl] #477641
07/18/19 12:30
07/18/19 12:30
Joined: Jul 2017
Posts: 783
Z
Zheka Offline OP
User
Zheka  Offline OP
User
Z

Joined: Jul 2017
Posts: 783
1) It's not a bad idea, its how the whole industry defines and calculates it. There is no other software nor paper that calculates it differently. That's why I raised the question in the first place.

2) Reporting "MaxDD in %" for non-reinvesting at all IS a bad idea, exactly for the reasons you described.
What IS used is the inverse of your maxDD/(ending)NP calculation - an NP/maxDD ratio (and preferably annualized) - a reward for assumed risk.

Now, it is your product and you can calculate it however you deem appropriate. Just make a note in bold and in capital in the manual that this deviates from the usual calculation which users take for granted.


And if there are issues "of opinion", why not have a vote/discussion on the forum - or- better - a 'pushed' user poll by email ?

Last edited by Zheka; 07/18/19 12:46.
Page 1 of 2 1 2

Moderated by  Petra 

Powered by UBB.threads™ PHP Forum Software 7.7.1