Re: st: Multiple imputation with panel data

[Oliver Jones <ojones@wiwi.uni-bielefeld.de>]

Hi Veronica,

I was suggesting a way how to get resonable values for x2007. But as I said, these estimated 
values for x2007 don't help you in explaining y2007 (which I assume is your ultimate goal?!), 
because they do NOT contain any new information.


Am 06.07.2012 11:33, schrieb Veronica Galassi:
> So maybe I should first try to estimate the coefficients of x2003 = b_0 +
> b_1*y2003 and then use them to predict x2007 as Oliver was suggesting.

If you call linear regression where I assume that the coefficient doesn't change over time 
(i.e. it's the same linear relationship between x and y in both years) "extrapolating values 
for x2007 exploiting the linear relationship between x and y" then yes, that is what I was 
talking about. But again: It just gives your estimates for the missing x2007 but dose NOT help 
explaining y2007!

> But isn't this way of proceeding the same than extrapolating values for
> x2007 exploiting the linear relationship between x and y?
> Because maybe I could simply use extrapolation.


Best Oliver
*
*   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/