PERFORMANCE METRICS IN PERFORMANCE REPORTS

Questions about MultiCharts and user contributed studies.
khalaad
Posts: 312
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 62 times
Been thanked: 57 times

PERFORMANCE METRICS IN PERFORMANCE REPORTS

Postby khalaad » 18 Apr 2012

A few days ago I ran a portfolio backtest on data starting January 2012. The report returned some astronomical Calmer numbers.

I fail to understand this as Calmer Ration requires at least 3 years of data, and a 3.0 Clamer is outstanding. None of the numbers I got were lower than 5.

The report is 31MB; therefore, not attached.

Khalid

khalaad
Posts: 312
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 62 times
Been thanked: 57 times

Re: PERFORMANCE METRICS IN PERFORMANCE REPORTS

Postby khalaad » 20 Apr 2012

Would not MultiCharts please care to respond?

I think this is an important matter of allocating a user's dollars on the basis of reports generated by the software.

Thank you.

Khalid

User avatar
JoshM
Posts: 2080
Joined: 20 May 2011
Location: The Netherlands
Has thanked: 1525 times
Been thanked: 1483 times
Contact:

Re: PERFORMANCE METRICS IN PERFORMANCE REPORTS

Postby JoshM » 20 Apr 2012

khalaad wrote:Would not MultiCharts please care to respond?

There's indeed not a lot of activity from MC support on the forum lately, but according to MC support earlier this week..
..we have been receiving a higher than usual number of inquiries.

I suppose they're all working on the release of MC 8 Release Candidate. :D

I do wonder if TJ is on a vacation, since I haven't seen a post from him in a while here (or I looked in the wrong places :) ).

On-topic: I suppose the Performance Report annualizes the return and then calculates the Calmer ratio on this, but I can't prove this assumption mathematically.

Edit: According the the Performance Report, all available data is used when there is less than 3 years available:
calmarRatio.PNG
calmarRatio.PNG (61.36 KiB) Viewed 355 times


So, I suppose that means the Calmar Ratio can only be used when you have at least three years of data. And if you have less than 3 year, it uses the return over that period as a substitute for the 3 year return.

khalaad
Posts: 312
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 62 times
Been thanked: 57 times

Re: PERFORMANCE METRICS IN PERFORMANCE REPORTS

Postby khalaad » 20 Apr 2012

Well, then this should be called MAR Ratio.

http://en.wikipedia.org/wiki/Calmar_ratio

Khalid
These users thanked the author khalaad for the post:
JoshM

khalaad
Posts: 312
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 62 times
Been thanked: 57 times

Re: PERFORMANCE METRICS IN PERFORMANCE REPORTS

Postby khalaad » 20 Apr 2012

More on MAR Ratio, by Joe Ross:

I always felt the single most cause for investors pain is unrealistic expectations. Investing/finance is the only field in the world where everyone is a genius. At parties, I don't even tell people what I do. I tell them I'm a bricklayer. If I mention I'm in finance, or trading, guaranteed, within 45 seconds they're telling me what I should buy or sell. Ever meet a brain surgeon and then assume you know more than him about brain surgery? Only in finance. Nope, when I'm socializing I'm a bricklayer. But here's one thing this bricklayer knows. Unrealistic expectations will blow you up. So what's realistic. A quick, easy metric to use is called a MAR Ratio or Managed Accounts Reports Ratio. It is simply the average compounded return divided by the one time worst drawdown. Translated into, the price you must pay to earn those returns in a specific strategy. So what's a good MAR ratio. Buffett, Dennis, Eckhardt, Neff, Dolphin, Saxon; all 'Hall of Fame' managers have a MAR ratio less than 1/2. So if you're shooting for a 20% return per year, well you can absolutely count on at least 40% drawdown – but only if you're as good as the above list. You need a good five years or so to get a ratio that means anything. The law of large numbers always results in some clown making 60% returns for a year or two only to get blown up the next. The only exception I noticed is option writing strategies. They can produce MAR ratio's of many multiples for 5–6 years, but then they usually blow up. To make money in those types of strategies you really have to toe the line, and eventually, well, you fall over it. So if you see some extraordinary returns with little or no risk, well, maybe you have the Midas gold touch, but then again maybe it's just fools gold. Caveat emptor!

User avatar
TJ
Posts: 6551
Joined: 29 Aug 2006
Location: Global Citizen
Has thanked: 966 times
Been thanked: 1892 times

Re: PERFORMANCE METRICS IN PERFORMANCE REPORTS

Postby TJ » 20 Apr 2012

JoshM wrote:...
I do wonder if TJ is on a vacation, since I haven't seen a post from him in a while here (or I looked in the wrong places :) ).
....

I only answer the easy questions.
I leave the difficult ones to JoshM et al.
...and the questions have been difficult lately.

;-)

User avatar
JoshM
Posts: 2080
Joined: 20 May 2011
Location: The Netherlands
Has thanked: 1525 times
Been thanked: 1483 times
Contact:

Re: PERFORMANCE METRICS IN PERFORMANCE REPORTS

Postby JoshM » 20 Apr 2012

TJ wrote:I only answer the easy questions.
I leave the difficult ones to JoshM et al.
...and the questions have been difficult lately.

;-)

Haha :D How calculated of you. ;)

Have a nice weekend!
These users thanked the author JoshM for the post:
TJ

User avatar
Henry MultiСharts
Posts: 8443
Joined: 25 Aug 2011
Has thanked: 1207 times
Been thanked: 2704 times

Re: PERFORMANCE METRICS IN PERFORMANCE REPORTS

Postby Henry MultiСharts » 11 Oct 2013

Calmar ratio = (minimum acceptable rate of return/12) / AbsValue (Max strategy drawdown %).
Minimum acceptable rate of return is user specified in Strategy performance report->Settings->Financial (please see calmar1.png).
You need to get Max strategy drawdown % from Periodical analysis->Monthly returns and drawdowns (%)-the lowest value (calmar2.png)
For example Minimum acceptable rate of return =2.
Max strategy drawdown = -0.00319
Calmar ratio=(2/12) / 0.00319=0.1666666666666667/0.00319=52.24660397074191 (calmar3.png)
Attachments
calmar1.png
calmar1.png (103.86 KiB) Viewed 342 times
calmar2.png
calmar2.png (146.22 KiB) Viewed 343 times
calmar3.png
calmar3.png (75.35 KiB) Viewed 344 times


Return to “MultiCharts”