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

RE: st: RE: Variogram for longitudinal data

From   "Nick Cox" <>
To   <>
Subject   RE: st: RE: Variogram for longitudinal data
Date   Tue, 27 Oct 2009 15:27:51 -0000

I guess that they haven't written it for precisely the reasons you
haven't written it. Most people prefer that someone else does it. (Me
too; I really wouldn't have the time, inclination or ability to write my
own text editor, mailer or Mata compiler.) 

After all, why there isn't there a Stata program for anything else you
want to name that hasn't been tackled already? 


Raphael Fraser

I am very surprize that nobody as wrote a variogram program for
longitudinal data since its such an important part of analyzing
longitudinal data. Perhaps individuals just wrote their own programs
which they have not 'shared." I think I will try to clone variog2.

Thanks to Nick and Stas for responding.

On Tue, Oct 27, 2009 at 8:57 AM, Nick Cox <> wrote:

> The programs -variog- and -variog2- are in the -variog- package on
> 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
> -variog2- (the more relevant here) pays no attention to the special
> structure of longitudinal data but unless otherwise instructed lumps
> panels together. (You could use -variog2- with panel data by
> 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
> 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
> 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:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index