PureBytes Links
Trading Reference Links
|
Dingo,
CAR is simply Cumulative Annual Return and is identical the the
Annual % return in AB
Max System DD is the one that's important to me.
MAR is simply CAR / MDD.
Regarding optimization with more then two parameters, the problem of
course is that one would require a more than 3 dimensional plot. and
obviously this is not possible. However there are ways around some
this. for example if your are using the following 3 variables
X = 1 to 10 by 1
Y = 1 to 10 by 1
Z = 1 to 10 by 1
Then one could combine the X & Y values which would create 100 pairs
for one axis etc. but this is not always achievable or readily
viewable in a 3d plot and as a result sometimes it's just best to
export the result, bring it in Excel, create a column for MAR, sort
by MAR and then lop off the bottom of the results that are say not in
the top 5 or 10 or 20% and then look at what's left.
Fred
--- In amibroker@xxxxxxxxxxxxxxx, "dingo" <dingo@xxxx> wrote:
> Fred,
>
> can you point me to where your MAR, CAR, etc are defined? As to
DD's
> there are 2 provided by the optimizer: system and max system which
one
> are you referring to?
>
> As to optimizing: what if you've got more than 2 variables (say 5
or 6)
> and were interesting in using that contour plot so you can see the
best
> peaks and miss the canyons. The contour plot (as I remember) allows
only
> 2 variables so how do you use it? IOW which 2 params would you pick?
>
> TIA
>
> d
>
> -----Original Message-----
> From: Fred <fctonetti@xxxx> [mailto:fctonetti@x...]
> Sent: Thursday, February 06, 2003 1:54 PM
> To: amibroker@xxxxxxxxxxxxxxx
> Subject: Stock Selection was: [amibroker] Re: NDX / QQQ - Can it be
> traded ?
>
>
> Al,
>
> Although I won't disagree with this approach as I've used it
myself,
> I will also tell you that it does have the potential for missing
the
> most robust area.
>
> If for example in using the 25 to 200 by 25 apporach it turns up
100
> as having the best results one would then be tempted to optimize
from
> 75 to 125 by some smaller increment say 5. However because of the
> span of the original increment ( 25 ) the most robust area may in
> fact be "hiding" between 53 and 71 which of course gets totally
> missed. These things become a tradeoff between available computer
> time and other priorities, but I will tell you I do upon occasion
set
> up optimizations in the evening knowing that they won't finish
until
> the following morning.
>
> Fred
>
> --- In amibroker@xxxxxxxxxxxxxxx, "Al Venosa" <avcinci@xxxx> wrote:
> >
>
>
>
> Yahoo! Groups Sponsor
>
> ADVERTISEMENT
>
>
<http://rd.yahoo.com/M=243066.2784922.4151385.1927555/D=egroupweb/S=17
05
> 632198:HM/A=1377502/R=0/*http://www.verisign.com/cgi-bin/go.cgi?
a=b31550
> 113206004000>
>
> <http://us.adserver.yahoo.com/l?
M=243066.2784922.4151385.1927555/D=egrou
> pmail/S=:HM/A=1377502/rand=730268776>
>
> 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 the Yahoo! Terms of Service
> <http://docs.yahoo.com/info/terms/> .
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/
|