Re: Fwd: Question about regCoef@rstd

I inadvertently did not post the response to ncl-talk

Date: Mon, 31 Jan 2011 08:10:24 -0700
Subject: Re: Question about regCoef@rstd

The routine is based on
c . see: brownlee
c . statistical theory and methodology
c . j wiley 1965 pgs: 281-284 QA276 .B77
=============================
To avoid semantic misunderstanding, here is a
snippet of the fortran code:

[snip]
         xave = sngl(xsum)/xyn
         yave = sngl(ysum)/xyn

         xvar = x2sum-xsum*xsum/dble(xyn)
         yvar = y2sum-ysum*ysum/dble(xyn)
         xyvar = xysum-xsum*ysum/dble(xyn)

         rcoef = sngl(xyvar/xvar) ! regression coef (b in book)
         ssqreg = xyvar*(xyvar/xvar) ! sum of squares due to
regression
         ssq = (yvar-ssqreg)/dble(xyn-2.)
         vb = ssq/xvar ! v[b] in book {variance
of B}
         sqrtvb = dsqrt(vb)
         tval = (rcoef-rnull)/sngl(dsqrt(vb))! t-statistic
         df = xyn-2. ! degrees of freedom

[snip]

"sqrtvb" is returned as the "rstd" attribute

Good luck

> Begin forwarded message:
>
>> *From: *Yohan Ruprich-Robert <Yohan.Ruprich-Robert@cerfacs.fr
>> <mailto:Yohan.Ruprich-Robert@cerfacs.fr>>
>> *Date: *January 31, 2011 2:58:15 AM MST
>> *To: *ncl-talk@ucar.edu <mailto:ncl-talk@ucar.edu>
>> *Subject: **[ncl-talk] Question about regCoef@rstd*
>>
>> Hi All,
>>
>> I would like to know what exactly is the rstd attribute of the function
>> regCoef:
>>
>> Is it the RMSE between the regression curve at each points on the
>> regression plot ?
>> Or is it somthing else ?
>>
>> Thanks in advance for your help,
>> Yohan RUPRICH
>> _______________________________________________
>> ncl-talk mailing list
>> List instructions, subscriber options, unsubscribe:
>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>

-- 
======================================================
Dennis J. Shea                  tel: 303-497-1361    |
P.O. Box 3000                   fax: 303-497-1333    |
Climate Analysis Section                             |
Climate & Global Dynamics Div.                       |
National Center for Atmospheric Research             |
Boulder, CO  80307                                   |
USA                        email: shea 'at' ucar.edu |
======================================================
_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk