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/