Ok -- I'm not sure why you are interested in count. If you want to identify weekends & cumulate the incidence numbers, then you can do something along the following lines.
First, put your date in Stata elapsed date form (check out the date(s1,s2) function).
Second, generate the year, week & day of week using other Stata functions (namely year(), week() & dow())
Only keep those observations for the weekends (keep if dow == 5 | dow == 6)
Sort the data by year & week
And finally
collapse (sum) incid, by(year week)
Please note in your example below that May 6, 2000 falls on a Saturday, not a Friday. Also, this will cumulate the incid values for all Fridays & Saturdays of a given year & week. In other words, the program doesn't care whether you have Friday only, or say five observations for a given Saturday.
Hope this helps,
Eric
>I have data of the form:
>
>date dw incid
>6/may/00 friday 5
>7/may/00 saturday 9
>13/may/00 friday 10
>14/may/00 saturday 12
>
>dw is day of week and incid is the number of violent incidents recorded in police stats.
>
>I wanted to collapse by weekend (here assuming a weekend is a Friday and Saturday) and therefore added in the following id (count):
>
>gen count = 0
>
>local cntr = 1
>forval i = 1(2)`=_N' {
>qui replace count = `cntr' in `i'
>local cntr = `cntr' + 1
>}
>local cntr = 1
>forval i = 2(2)`=_N' {
>qui replace count = `cntr' in `i'
>local cntr = `cntr' + 1
>}
>
>which gives:
>
>date dw incid count
>6/may/00 friday 5 1
>7/may/00 saturday 9 1
>13/may/00 friday 10 2
>14/may/00 saturday 12 2
>
>and then:
>
>collapse (sum) incid (min) date, by(count)
>
>I have two questions. First, I am certain I have seen a more elegant way of creating the count id posted on Statalist but searching has not revealed anything. Is there one?
>
>More importantly, if, for some reason, there is a missing saturday in the data (there isn't at the moment but it is a growing dataset) the collapse ..., by(count) will not work. Is there an easier way of numbering successive weekends in Stata?
>
>Regards
>
>Simon Moore
>*
>* 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/
--
===================================================
Eric G. Wruck
Econalytics
2535 Sherwood Road
Columbus, OH 43209
ph: 614.231.5034
cell: 614.327.4182
fax: 614.231.5039
eFax: 614.573.6639
eMail: [email protected]
website: http://www.econalytics.com
====================================================
*
* 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/
