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

[amibroker] Re: A technical question



PureBytes Links

Trading Reference Links

Equal weighting is more robust and a better choice in most cases. 
Either a factor is good enough to use or it isn't. You can find
weights that give better results in hindsight through brute force
optimization but in reality the "best weights" are dynamic.

--- In amibroker@xxxxxxxxxxxxxxx, "Herman vandenBergen" <psytek@xxxx>
wrote:
> MessagePaul,
> 
> When you include "different" factors in a PositionScore formula they
may not
> be naturally weighted equally, so using simple Sums or Products
formulas may
> not always give you the best results.
> 
> Take a Sum example: PositionScore = FundamentalFactor + PriceFactor;
> this assumes both terms carry equal weight in terms of
PricePrediction. Why
> would this be so?
> 
> Most likely you can optimize the formula by adding a Weight
adjustment
> factor:
> PositionScore = Optimize("WAF",1,0,2,0.1)*FundamentalFactor +
PriceFactor;
> This will gradually shift the weight from the left to the right
term, try
> different opt ranges.
> 
> When your equation is a Product equation then scaling one or the
other will
> not effect your score as the ratio between the two remains constant
by
> multiplication, you would just be scaling the overall result.
> In this case you can Optimize the weighting of the terms using
powers:
> PositionScore = FundamentalFactor^Optimize("WAF",1,0,2,0.1) *
PriceFactor;
> 
> If your scoring formula is complex you may have to try a
combination of
> these weight adjustment techniques.
> 
> Good luck,
> herman.
> 
>   -----Original Message-----
>   From: Paul Ho [mailto:paultsho@x...]
>   Sent: January 19, 2004 10:19 AM
>   To: amibroker@xxxxxxxxxxxxxxx
>   Subject: RE: [amibroker] Re: Question on Position Score ranking
> 
> 
>   Al, I have tried it but I dont have any definitive conclusion so
will be
> interesting to see your results, you can trying multiplying as well
as
> additions. Would you mind let me know once you have completed your
exercise.
> I have tried it on a breakout system. I figured that I want both
random()
> and NV2 have the right degree of influence.
> 
>  
NV2=(V-BBandBot(V,50,2))/(0.000000001+BBandTop(V,5
0,2)-BBandBot(V,50,2));
>   PositionScore=Random()+NV2;
>   // or this one PositionScore=Random()*NV2;
> 
>   /Paul.
>     -----Original Message-----
>     From: Al Venosa [mailto:advenosa@x...]
>     Sent: Monday, 19 January 2004 12:52 PM
>     To: amibroker@xxxxxxxxxxxxxxx
>     Subject: Re: [amibroker] Re: Question on Position Score ranking
> 
> 
>     Thank you, Paul. Believe it or not, I just came to that very
conclusion
> as I was testing. You are indeed correct. It's probably best simply
to use
> PositionScore = Random(), which simply picks any 5 stocks randomly
from the
> watchlist. That way, one can compare randomly picked stocks with
stocks
> having high volume to evaluate the volume variable in its purest
form. A
> very neat little trick.
>       ----- Original Message -----
>       From: Paul Ho
>       To: amibroker@xxxxxxxxxxxxxxx
>       Sent: Sunday, January 18, 2004 8:46 PM
>       Subject: RE: [amibroker] Re: Question on Position Score
ranking
> 
> 
>       Al,
>       random() generates number between 0 and 1 and in order to be
> effective, you'll have to normalise your volume or even better still
express
> it in terms of standard deviation such as applying bollinger bands
to it. so
> the volume number is within the same kind of range. otherwise, a
nominally
> higher volume stock will have too much bias.
>       Paul.
>         -----Original Message-----
>         From: Al Venosa [mailto:advenosa@x...]
>         Sent: Monday, 19 January 2004 11:48 AM
>         To: amibroker@xxxxxxxxxxxxxxx
>         Subject: Re: [amibroker] Re: Question on Position Score
ranking
> 
> 
>         Hmm. What an interesting idea. So, in my case, my
positionscore line
> would
>         be:
> 
>         PositionScore = Random()*V;
> 
>         Thanks for the tip. At least I can evaluate how good the
ranking
> system is
>         working. But in order to actually select the second 5 top
stocks, I
> guess I
>         still have to use the Osaka plugin as Fred pointed out.
> 
>         Thanks a lot, Thelunit.
> 
>         Al V.
> 
> 
> 
> 
>         ----- Original Message -----
>         From: "thelunit" <thelunit@xxxx>
>         To: <amibroker@xxxxxxxxxxxxxxx>
>         Sent: Sunday, January 18, 2004 7:08 PM
>         Subject: [amibroker] Re: Question on Position Score ranking
> 
> 
>         > >>or maybe I just want to see how well my ranking formula
picks
> winners)
>         > <<
>         >
>         >
>         > Here's one idea..
>         >
>         > I was interested in considering the effectiveness of my
> 'PositionScore'
>         specification a little while back.
>         >
>         > What I did was run repeated dummy optimizations with
random
> ranking to
>         > generate a distribution of system stats against which I
ccould
> compare the
>         stats produced by my non-random PositionScore (in your case
for eg,
> V).
>         >
>         > These lines give the 'null' reference distribution:
>         >
>         > PositionScore = Random();
>         > dummyopt = Optimize("dummyopt", 1, 1, reps, 1); // reps =
your
> preference
>         >
>         > Best
>         >
>         >
>         >
>         > --- In amibroker@xxxxxxxxxxxxxxx, "Al Venosa"
<advenosa@xxxx>
> wrote:
>         > > Hi, all:
>         > >
>         > > I have a question on positionscore in regular mode.
Suppose I'm
> trading
>         from a watchlist of, say, 100 stocks (like the N100, for
examplem,
> but it
>         can be anything) and my portfolio size is 5 stocks
(positionsize
>         = -100/posqty, where posqty = 5). And suppose I want to
select the
> highest
>         volume stocks to trade, like:
>         > >
>         > > PositionScore = V;
>         > >
>         > > Now, my question is, what if I wanted to trade, not the
TOP 5
> stocks on
>         the rank list (stocks with rankings of 1st, 2nd, 3rd, 4th,
and 5th
> in terms
>         of volume), but rather the next 5 stocks (ranks of 6th, 7th,
8th,
> 9th, and
>         10th). My reasons for wanting to do this are not important
(maybe I
> don't
>         want to compete for the same 5 stocks as another guy who
uses the
> same
>         ranking formula, or maybe I just want to see how well my
ranking
> formula
>         picks winners). How would you code that? Any ideas?
>         > >
>         > > Thanks.
>         > >
>         > > Al Venosa
>         > > advenosa@xxxx
>         >
>         >
>         >
>         > Send BUG REPORTS to bugs@xxxx
>         > Send SUGGESTIONS to suggest@xxxx
>         > -----------------------------------------
>         > Post AmiQuote-related messages ONLY to:
amiquote@xxxxxxxxxxxxxxx
>         > (Web page: http://groups.yahoo.com/gro
up/amiquote/messages/)
>         > --------------------------------------------
>         > Check group FAQ at:
>         http://groups.yahoo.com/group/amibroker/files/groupfaq.html
>         >
>         >
>         > Yahoo! Groups Links
>         >
>         > To visit your group on the web, go to:
>         >  http://groups.yahoo.com/group/amibroker/
>         >
>         > To unsubscribe from this group, send an email to:
>         >  amibroker-unsubscribe@xxxxxxxxxxxxxxx
>         >
>         > Your use of Yahoo! Groups is subject to:
>         >  http://docs.yahoo.com/info/terms/
>         >
>         >
> 
> 
>         ---
>         Outgoing mail is certified Virus Free.
>         Checked by AVG anti-virus system (http://www.grisoft.com).
>         Version: 6.0.561 / Virus Database: 353 - Release Date:
1/13/2004
> 
> 
> 
>         Send BUG REPORTS to bugs@xxxx
>         Send SUGGESTIONS to suggest@xxxx
>         -----------------------------------------
>         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
> 
> 
> 
> ------------------------------------------------
------------------------
>         Yahoo! Groups Links
> 
>           a.. To visit your group on the web, go to:
>           http://groups.yahoo.com/group/amibroker/
> 
>           b.. To unsubscribe from this group, send an email to:
>           amibroker-unsubscribe@xxxxxxxxxxxxxxx
> 
>           c.. Your use of Yahoo! Groups is subject to the Yahoo!
Terms of
> Service.
> 
> 
> 
> 
>       Send BUG REPORTS to bugs@xxxx
>       Send SUGGESTIONS to suggest@xxxx
>       -----------------------------------------
>       Post AmiQuote-related messages ONLY to: am
iquote@xxxxxxxxxxxxxxx
>       (Web page: http://groups.yahoo.com/group/amiquote/messages/)
>       --------------------------------------------
>       Check group FAQ at:
> http://groups.yahoo.com/group/amibroker/files/groupfaq.html
> 
> 
> 
> 
>
--------------------------------------------------
------------------------
>       Yahoo! Groups Links
> 
>         a.. To visit your group on the web, go to:
>         http://groups.yahoo.com/group/amibroker/
> 
>         b.. To unsubscribe from this group, send an email to:
>         amibroker-unsubscribe@xxxxxxxxxxxxxxx
> 
>         c.. Your use of Yahoo! Groups is subject to the Yahoo!
Terms of
> Service.
> 
> 
> 
> 
>     Send BUG REPORTS to bugs@xxxx
>     Send SUGGESTIONS to suggest@xxxx
>     -----------------------------------------
>     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
> 
> 
> 
>
--------------------------------------------------
--------------------------
>     Yahoo! Groups Links
> 
>       a.. To visit your group on the web, go to:
>       http://groups.yahoo.com/group/amibroker/
> 
>       b.. To unsubscribe from this group, send an email to:
>       amibroker-unsubscribe@xxxxxxxxxxxxxxx
> 
>       c.. Your use of Yahoo! Groups is subject to the Yahoo! Terms
of
> Service.
> 
> 
> 
> 
>   Send BUG REPORTS to bugs@xxxx
>   Send SUGGESTIONS to suggest@xxxx
>   -----------------------------------------
>   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
> 
> 
> 
>
--------------------------------------------------
--------------------------
> --
>   Yahoo! Groups Links
> 
>     a.. To visit your group on the web, go to:
>     http://groups.yahoo.com/group/amibroker/
> 
>     b.. To unsubscribe from this group, send an email to:
>     amibroker-unsubscribe@xxxxxxxxxxxxxxx
> 
>     c.. Your use of Yahoo! Groups is subject to the Yahoo! Terms of
Service.


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 

Yahoo! Groups Links

To visit your group on the web, go to:
 http://groups.yahoo.com/group/amibroker/

To unsubscribe from this group, send an email to:
 amibroker-unsubscribe@xxxxxxxxxxxxxxx

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