Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: How to replace missing values in panel data


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: How to replace missing values in panel data
Date   Wed, 20 Oct 2004 15:28:37 +0100

-ipolate- does linear interpolation, as its help 
explains. What would seem to make most sense 
with panel data is to interpolate with respect 
to time. As -igrs- appears not to be your time variable, 
your -ipolate- call might indeed produce results you 
don't expect. You should probably use -year- 
instead. Naturally, the explicit assumption here 
is of linear segments between non-missing values. 

-impute- uses a regression method, on which 
see [R] impute. 

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

soma ghosh

> I have a data set from 1977 to 1992 for every 5 yrs
> (77, 82, 87, 92) with some missing values for my
> dependent variables. For e.x. data on expenditure is
> missing for the year 1977 and there are also some
> missing values in the other years. I tried using both
> impute and ipolate command in stata to get values for
> 1977. I used 
> sort id year
> by id: ipolate exp igrs, gen (iexp) epolate.
> 
> But quite a few of the resulting values (about 30%)
> are negative & the ipolated values don't seem to be
> correct (way different from expenditures in other
> years). Can anybody help me to find a correct way to
> do this? Also, what is the mechanism behind ipolate or
> impute? 

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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