<>
Just wondering: Would the -tsfill- line (which adds observations for every
single day spent by every single patient in the dataset) lead to memory
problems at some point, given that Rajaram said that sample members spent
two years there in some cases?
HTH
Martin
-----Ursprüngliche Nachricht-----
Von: [email protected]
[mailto:[email protected]] Im Auftrag von Kit Baum
Gesendet: Montag, 2. November 2009 17:06
An: [email protected]
Betreff: st: re: Creating number of days per month
<.>
It was suggested that the long format makes this easy: to compute how
many days in each calendar month the patient was in the facility.
Indeed so:
clear
inp str10(var1 var2)
"30nov2007" "14feb2008"
"21jan2009" "07apr2009"
"24apr2008" "09jul2008"
"22feb2008" "08may2008"
"01oct2008" "16dec2008"
"13jul2009" "28sep2009"
"04oct2008" "20dec2008"
"07jul2009" "22sep2009"
"03jun2008" "19aug2008"
"11feb2009" "30apr2009"
"31mar2007" "17jun2007"
"01may2007" "19jul2007"
"04apr2007" "22jun2007"
"03jul2008" "20sep2008"
"10jul2007" "27sep2007"
end
gen date1=date(var1, "DMY")
gen date2=date(var2, "DMY")
drop var1 var2
g patientid = _n
reshape long date, i(patientid)
tsset patientid date
format date %td
l, sepby(patientid)
tsfill
g mon = mofd(date)
g days = 1
collapse (count) days, by(patientid mon)
format mon %tm
l patientid mon days, sepby(patientid)
Kit Baum | Boston College Economics & DIW Berlin |
http://ideas.repec.org/e/pba1.html
An Introduction to Stata Programming
| http://www.stata-press.com/books/isp.html
An Introduction to Modern Econometrics Using Stata |
http://www.stata-press.com/books/imeus.html
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
