Re: Lag regression...

On 07/12/2010 10:46 AM, Sabeerali(sebi) wrote:
> Does any function in NCL to compute lag regression just like lag
> correlation...?

No.

> Any hint would be highly appreciated...

Shift the time axis manually.

For the simple x(N), y(N)

    N = 10
    x = random_normal(0.0,1.0,N)
    y = random_normal(0.0,1.0,N)

     nlagr = 3 ; 0,1,2
     rlag = new (nlagr, "float")
     Nend = N-1
     yLag = 1
     do n=0,nlagr-1
        rlag(n) = regline(x(0:Nend-n-yLag) , y(yLag+n:Nend) )
     end do
     print(rlag)

===========
For multidimensional arrays use

http://www.ncl.ucar.edu/Document/Functions/Built-in/regCoef.shtml
_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Mon Jul 12 12:04:58 2010