If you map your years 1991/1992 1994 on to times 1, 2, 3 on then you can
get Stata to -tsset- using those times. But you would be pretending that
a gap does not exist. Yet the alternative is an enormous block of
missings.
The choice is yours, and the consequences are yours, not Stata's.
Nick
[email protected]
Viktor Slavtchev
Maarten, you are right I should clarify the point.
There is yearly questionnaire I am relying on. However, for some reason,
this questionnaire was not performed in 1993. That is, the 1993 wave
simply dies not exist. The data for 1991, 1992, 1994 and so on are
available. My question is, whether this is problem in exploiting the
panel structure of the data and how to deal with?
Maarten buis wrote:
> --- Viktor Slavtchev <[email protected]> wrote:
>
>> I have a panel with one missing year. For example the data is
>> available for 1991, 1992, 1994, 1995 and so on, but not for 1993.
>> That is not because the survey, I've got the data from, was not
>> performed in the year 1993. My question is how to use the panel
>> structure of the data, considering that particular break in the data?
>>
>
> Do I understand you correctly that the survey was performed in that
> year but that year is missing for everyone? That to me seems to
suggest
> a serious problem in the data preparation (e.g. you appended the
> different waves to create a single file, but somehow the append for
> wave 1993 failed), and I would go back to that before trying to do any
> analysis.
*
* 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/
