If you are not sure about whether each year is represented:
* find problems
gen elapsed_years = (year - year[_n-1]) if (zip == zip[_n-1])
expand elapsed_years
sort zip year
* replace the now bogus counts
replace count = 0 if ((elapsed_years > 1) & (year == year[_n-1]) & (zip == zip[_n-1]))
* get the year right
replace year = (year[_n-1] + 1) if (zip == zip[_n-1])
This does assume that the first year and last year for each zip code are observed with non zero counts. Otherwise you would not be able to tell it is missing from the setup of the data.
-Tim
-----Original Message-----
From: Nick Cox [mailto:[email protected]]
Sent: Tuesday, May 25, 2004 4:49 PM
To: [email protected]
Subject: st: RE: "imputing" missing observations
So long as 1989 is a value for some other
zip, you can
fillin zip year
replace count = 0 if _fillin
Nick
[email protected]
Heather Taffet Gold, PhD
> I've been looking all around, asking around, etc, and I can't
> seem to find
> out how to create new observations within my dataset.
>
> Here are the specifics of my problem:
> I have counts of cases by zip code and year (one observation
> with count for
> each zip, and year). It happened that I ended up with "missing"
> observations if the count of cases was zero in a year, so my
> data look
> something like this
> zip year count
> 06010 1986 1
> 06010 1987 2
> 06010 1988 2
> 06010 1990 7
> ...
>
> How would I create a new observation for this zipcode in 1989
> that has a
> count of 0? I need to be able to create some lagged variables
> by year and zip.
*
* 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/
*
* 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/
