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

Re: [amibroker] Re: Polynomial fit



PureBytes Links

Trading Reference Links

Aequalsz:

It sounds like you have made some good progress.  Would it be possible for
you to either post or send me the source code for the dll?

Bill


----- Original Message ----- 
From: "aequalsz" <aequalsz@xxxxxxxxx>
To: <amibroker@xxxxxxxxxxxxxxx>
Sent: Tuesday, September 23, 2003 12:55 PM
Subject: [amibroker] Re: Polynomial fit


--- In amibroker@xxxxxxxxxxxxxxx, "wavemechanic" <wd78@xxxx> wrote:
> Stephane:
>
> I am not referring to a smoothing function (e.g., Gaussian) but
rather
> fitting data(x, y) to a polynomial equation of degree k.
Essentially what
> the POLYNOMIAL function in Excel does.  In that case, Excel uses a
least
> squares fit of the data to a polynomial of degree k.
> Bill
>
> ----- Original Message ----- 
> From: "Stephane Carrasset" <nenapacwanfr@xxxx>
> To: <amibroker@xxxxxxxxxxxxxxx>
> Sent: Tuesday, September 23, 2003 11:45 AM
> Subject: [amibroker] Re: Polynomial fit
>
>
> Hi bill,
>
> do you refer to :"
> The Gaussian average are computed according to Ehlers ( Easy
Language
> Code)
> One Pole: f = fÑg + (1-fÑ)f[1]
> Two Poles: f = fÑ2g + 2(1-fÑ)f[1] - (1-fÑ)2f[2]
> Three Poles: f = fÑ3g + 3(1-fÑ)f[1] - 3(1-fÑ)2f[2] +
(1-fÑ)3f[3]
> Four Poles: f = fÑ4g + 4(1-fÑ)f[1] - 6(1-fÑ)2f[2] +
4(1-fÑ)3f[3] - (1-
> fÑ)4f[4]
>
> or anything else?
>
> Stephane
>
> > Is anyone aware of code for producing a polynomial fit of price
> data, moving
> > average, etc. written in a translatable language (e.g.,
Metastock)?
> >
> > Thanks.
> >
> > Bill
>
>
>
> 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
>
> Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/

Aequalsz has a linear and quadratic least squares fit in MetaStock
which was implemented via the MS Developer's Kit in a C++ dll.

The linear ls is similar to what you see here and in MS, but the
quadratic ls contains quite a few more terms, so it's useful to cast
the problem in terms of matrix algebra.  You would then need access
to matrix routines like a solver, triple matrix products, etc.

Using the quadratic ls routine - so far at least - has not convinced
me that it yields a particularly good representation of a stock price
to forecast future prices.  And a higher degree polynominal would seem
even more "iffy" to me.

a



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 ---------------------~-->
ReplayTV: Control live television
Special Sale: 50% off ReplayTV
CNet Ranked #1 over Tivo!
http://us.click.yahoo.com/aUMW7B/A6qGAA/ySSFAA/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/