Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: Variogram for longitudinal data


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Variogram for longitudinal data
Date   Tue, 27 Oct 2009 13:57:28 -0000

The programs -variog- and -variog2- are in the -variog- package on SSC.
Please remember to specify where user-written stuff you refer to comes
from, as long requested on this list. 

I believe the short answer to this to be No. 

I also think that the statement is a little misleading. Say rather that
-variog2- (the more relevant here) pays no attention to the special
structure of longitudinal data but unless otherwise instructed lumps all
panels together. (You could use -variog2- with panel data by specifying
one of the coordinates as time and the other as a constant variable
constructed as a pacifier.) 

I think you have at least five possibilities: 

1. There may be such a package, but neither of us knows about it. 

2. You can apply -variog2- to each panel separately, if that is the
implication, and pool the answers. 

3. You can clone -variog2- and change the code so that it supports panel
structure. The code is neither long nor especially complicated. 

4. You can wait for the program author (me) to generalise the program,
which is on my to do list, although I doubt I will get to it for a few
months yet. 

5. Apply -variog2- regardless as above. 

Nick 
n.j.cox@durham.ac.uk 

Raphael Fraser

Are there any programs available to estimate variogram for
longitudinal data? Note that -variog- or -variog2- does not apply to
longitudinal data.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index