[EMAIL PROTECTED] wrote:
>
> Hi Edzer,
>
> Thanks for the clue.
> However, I am still confused and probably doing something wrong. Using the
> dX-option, I expect a reduction in the number of pairs for the semivariogram
> fitting. In my case with n=120 and three categories a reduction from
> 0.5*120*(120-1)= 7140 to 3*{0.5*40*(40-1)}=2340. Any suggestions?
>
If you're saying that not any reduction took place, please (1) check
your
X-columns thoroughly (they should be never equal), and if that does not
work, (2) please send the data as well, so I can give it a try.
--
Edzer
