>Delivered-To: shea@ucar.edu
>Delivered-To: ncl-talk@ucar.edu
>From: "ssarkar" <ssarkar@forestone.com>
>To: <ncl-talk@ucar.edu>
>Date: Mon, 10 Oct 2005 09:02:17 -0500
>
>Has anyone implemented a canonical correlation anaysis
>between a predictor and a predictand using NCL?
>If yes please let me know and any help in this
>regard will be very helpful
--- Ummm, seems like no one is going to answer ...[1] I have a fortran routine code that can be readily called from NCL. I have a simple example, if you are interested. It is *VERY* simple to call fortran subroutines from NCL. C is a bit more involved. [2] If you have available a library like NAG [Numerical Algorithms Group] or IMSL you can call the necessary routine via a simple fortran interface. [3] The "R" software package has a fumction call "cancor" which can be used. D
_______________________________________________ ncl-talk mailing list ncl-talk@ucar.edu http://mailman.ucar.edu/mailman/listinfo/ncl-talk
This archive was generated by hypermail 2b29 : Tue Oct 11 2005 - 19:09:25 MDT