Re: st: data manipulation

[Gary Longton <glongton@fhcrc.org>]

In addition to Nick's -reshape- solution, here's an alternative approach which 
uses -bysort- to fill in and drop observations:
The first 2 lines of the -forvalues- loop are just asserting what you imply is 
true - that there is indeed only one non-missing value for each series variable 
within country-year combination.

*****************************
forvalues i = 1/3 {
    egen notmiss = count(series`i'), by(country year)
    assert notmiss == 1
    qui bysort country year (series`i'): replace series`i' == series`i'[1]
    drop notmiss
}
qui by country year: drop if _n>1
*******************************

- Gary


jdubas@nd.edu wrote:


> Hello, 
> I have what may be a basic question.  I have data like this:
> country year series1 series2 series3
> 1          1      1000     .         .
> 1          2      1035     .         .
> 1         1        .        50        .
> 1          2       .         53       .
> 1         1        .          .        7 
> 1         2        .          .         8.5
> 2 etc
>
>
> Is there anyway I can basically collapse these so that:
> country  year  series1   series2  series3
> 1           1      1000       50        7
> 1           2      1035       53        8.5
>
> for a number of countries?


*
*   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/