Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Re: st: Missing data in panel analysis


From   Nick Cox <[email protected]>
To   [email protected]
Subject   Re: st: Missing data in panel analysis
Date   Thu, 7 Mar 2013 17:36:52 +0000

(previous was sent prematurely)

I doubt that there can be a general answer here that is correct across
all possibilities.

I didn't have an iPhone in 1987, so "No" rather than missing could be
a reasonable answer.

In any case, dropping an entire variable is not the only option. You
could -drop- the first record
for each panel.

Please use full real names on Statalist.

Nick

On Thu, Mar 7, 2013 at 5:21 PM, ARDE DE <[email protected]> wrote:

> In my panel analysis I have the following problem with missings:
>
> Assume that all individuals are observed throughout all time periods
> t_1, t_2, ..., t_T, for a certain set of variables. But a new variable
> is added at t_2 (and being answered for t_2, t_3, ..., by all
> individuals). Assume that this new variable did not make sense in t_1
> (e.g. do you have an iphone asked in 1987...) So, this variable is
> missing only for t_1.
> How can I deal with that problem?
> On the one hand I have missings in my matrix of regressors and cannot
> perform the usual inversions etc. On the other hand, it would be
> inefficient to drop the entire new variable.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index