Re: struct covariance question

[John Darrington]

I agree: There  ought to be such a function.  And no, it doesn't
exist right now.

Can you proceed by figuring it out yourself, or shall I hack up
something for you?

J'
 

On Tue, Feb 15, 2011 at 09:35:57PM -0500, Jason Stover wrote:
     To compute type 3 sums of squares, I need a submatrix
     of the covariance matrix, in which certain variables are
     dropped. Right now, I have this syntax:
     
     data list list / v0 to v2.
     begin data
     3.2 1 1
     3.1 1 1
     3.3 1 2
     3.4 1 2
     3.2 1 3
     3.3 1 3
     3.3 1 4
     3.2 1 4
     2.8 2 1
     2.9 2 1
     3.3 2 2
     3.0 2 2
     3.1 2 3
     3.2 2 3
     3.2 2 4
     3.1 2 4
     end data
     GLM v0 by v1 v2
         /INTERCEPT = include.
     
     This creates a 5-by-5 covariance matrix, since v1 and v2 are
     categorical. After running reg_sweep, I need to drop the rows and
     columns for v1 and re-run reg_sweep, then drop the rows and columns
     for v2 and re-run reg_sweep. I could do this by figuring out which
     columns these happen to be, but it seems there ought to be a place to
     do so in covariance.c.  Does covariance.c have such a function? I
     didn't see such a function there.
     
     The function I'm looking for would just tell me, say, which columns (or
     rows) correspond to v1, and which to v2.
     
     -Jason
     
     _______________________________________________
     pspp-dev mailing list
     address@hidden
     http://lists.gnu.org/mailman/listinfo/pspp-dev

-- 
PGP Public key ID: 1024D/2DE827B3 
fingerprint = 8797 A26D 0854 2EAB 0285  A290 8A67 719C 2DE8 27B3
See http://pgp.mit.edu or any PGP keyserver for public key.

signature.asc
Description: Digital signature