[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [amibroker] New portfolio testing numbers



PureBytes Links

Trading Reference Links

Hello,

Shortly on Standard Error figure:
standard error is a part of linear regression analysis.
For simple description see this:
http://www.numericalmathematics.com/least_squares.htm

Standard error of equity describes how much equity line
fluctuates around linear regression of it.

Note that standard error grows with your initial capital.
If the same system is tested usgin 10x more initial capital
then standard error will grow also 10x (assuming % commissions).

So you get high figures because your starting capital expressed
in YEN is high. You may prefer to divide standard error by starting capital
to get some more resonable values and to be able to compare
figures when you are using different initial capital.

Best regards,
Tomasz Janeczko
amibroker.com
----- Original Message ----- 
From: "Yuki Taga" <yukitaga@xxxxxxxxxxxxx>
To: <amibroker@xxxxxxxxxxxxxxx>
Sent: Sunday, October 05, 2003 8:54 AM
Subject: [amibroker] New portfolio testing numbers


> Here are of some of the new numbers that this new portfolio testing
> is throwing my way.  I do have a very rudimentary knowledge (stress
> on 'very') about some of these numbers, but certainly not a fraction
> as deep as some of you guys on this list.  Mainly, until this past
> year or so, I have been primarily a discretionary trader, one who has
> recently been looking seriously for a system, something that would
> attempt to trade systematically in a close approximation of what I
> have usually tried to do on a discretionary basis.
> 
> The data for this test goes back to 1/1/99, so it covers some strong
> bull, bear, and chop periods in Tokyo.  To me, the performance looks
> pretty darn good. The max. system DD is less than I expected (less
> than the number given by the old back tester, which means I'm getting
> some benefit from diversification), and easily tolerable. I'd sure
> like opinions on this, particularly on some of the numbers like UI
> and UPI and K-Ratio and Standard Error.  Recovery factor, for
> example, is something I think I might have a decent idea about, but I
> have never come across the actual statistical explanation.
> 
> So who would be inclined to use a system producing these numbers?
> Who would not? And why or why not, based on the performance numbers
> and about 5 years of data? Winners are 65.1 percent, long profit is
> about 52 percent of total versus about 48 percent for short profit.
> Clearly the long side is the better side over the test period, but
> the short side appears to me to be too good to ignore. (Long trade
> winners are about 36 percent of all trades, and short trade winners
> about 29 percent.) Testing this over 5 one-year periods shows very
> little variation in the numbers below, so I think they are
> representative of what I might expect.  But I really don't know the
> best way to determine that.
> 
> The size of the Standard Error numbers startles me.  I know standard
> deviation a bit, but not standard error, nor what it is actually
> telling me here.
> 
> Recovery Factor          16.42    12.68      4.57
> CAR/MaxDD                3.15     2.73       0.77
> RAR/MaxDD                24.59    35.94      8.43
> Profit Factor            2.41     3.08       1.94
> Payoff Ratio             1.29     1.36       1.27
> Standard Error           3112548.91          1569285.92     2911269.84
> Risk-Reward Ratio        4.62     5.29       2.09
> Ulcer Index              2.36     1.89       5.30
> UlcerPerformance Index   11.04    9.10       2.02
> Sharpe Ratio of trades   3.57     4.74       2.60
> K-Ratio                  6.34     7.25       2.87
> 
> Yuki
> 
> 
> 
> 
> Send BUG REPORTS to bugs@xxxxxxxxxxxxx
> Send SUGGESTIONS to suggest@xxxxxxxxxxxxx
> -----------------------------------------
> Post AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx 
> (Web page: http://groups.yahoo.com/group/amiquote/messages/)
> --------------------------------------------
> Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html 
> 
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/ 
> 
> 
> 

------------------------ Yahoo! Groups Sponsor ---------------------~-->
Buy Ink Cartridges or Refill Kits for your HP, Epson, Canon or Lexmark
Printer at MyInks.com. Free s/h on orders $50 or more to the US & Canada.
http://www.c1tracking.com/l.asp?cid=5511
http://us.click.yahoo.com/mOAaAA/3exGAA/qnsNAA/GHeqlB/TM
---------------------------------------------------------------------~->

Send BUG REPORTS to bugs@xxxxxxxxxxxxx
Send SUGGESTIONS to suggest@xxxxxxxxxxxxx
-----------------------------------------
Post AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx 
(Web page: http://groups.yahoo.com/group/amiquote/messages/)
--------------------------------------------
Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/