Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: RE: RE: "imputing" missing observations


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: RE: RE: "imputing" missing observations
Date   Tue, 25 May 2004 22:18:45 +0100

I think this can be approached also in 
an assumption-free way. 

1. use -edit- or -input- to add a block 
of observations with a bogus zip code and 
all years exemplified. 

2, 
fillin zip year 
replace count = 0 if _fillin 

3. 
delete the bogus observations


Nick 
n.j.cox@durham.ac.uk 

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu]On Behalf Of McKenna,
> Timothy
> Sent: 25 May 2004 22:11
> To: statalist@hsphsun2.harvard.edu
> Subject: st: RE: RE: "imputing" missing observations
> 
> 
> 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:n.j.cox@durham.ac.uk]
> Sent: Tuesday, May 25, 2004 4:49 PM
> To: statalist@hsphsun2.harvard.edu
> 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 
> n.j.cox@durham.ac.uk 
> 
> 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/
> 

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index