[ncl-talk] Significant test for Pattern Correlation

[Dennis Shea]

A **guess**

http://www.ncl.ucar.edu/Document/Functions/Built-in/rtest.shtml

uses


 t = *r**sqrt((*Nr*-2)/(1-*r^2*)

The key is estimating 'Nr'

Geophysical variables, generally, correlated in space
and time. The estimate of 'Nr' must take these issues into account.

Depending on the variable, the number of spatially
independent grid points may be significantly less than
the number of grid points.

Two papers:

http://www.atmo.arizona.edu/students/courselinks/fall09/atmo529/Lectures/Livezey_Chen_1983.pdf
 http://dx.doi.org/10.1175/1520-0442(1999)012<1280:EOSDOF>2.0.CO;2

You can do a google search for others.

====

Perhaps, you can use bootstrapping approach. You will have to

figue the correct approach on your own.




On Wed, Nov 2, 2016 at 6:58 AM, Xi Chang <xi.chang01 at gmail.com> wrote:

> Hallo,
>
> Is there any way to perform a significant test for
> a pattern correlation here http://www.ncl.ucar.edu/Document/Functions/
> Contributed/pattern_cor.shtml
>
> Thank you
> Chang
>
> _______________________________________________
> ncl-talk mailing list
> ncl-talk at ucar.edu
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20161102/030816de/attachment.html