More fundamentally, you would not gain by reducing
"3/12/03" to "31203".
Moreover, -destring- does not possess any special
features for recognising dates or times as such,
presumably on the grounds that other features
are available for that purpose.
Rather, I recommend that you take this
composite variable (-datetime-, say)
and -split- it first.
. split datetime
. gen date = date(datetime1,"dmy",2050)
. format date <whatever>
. drop datetime*
Nick
[email protected]
Eric Hixson
> I have a dataset with dates that I am brining into Stata
> via insheet. The data is comma delimited and the dates are
> being formatted by the source database like the following:
>
> 3/12/03 0:00:00
> 11/5/03 0:00:00
>
> Unfortunately I can't get rid of the h/m/sec portion at the
> source and Stata is formatting the field as a string. I
> want to convert the m/d/y portion to a date to allow time
> series analysis.
>
> I tried dstring as a first step and got the following:
>
> . destring dischdate, g(ddate) ignore(" 0:00:00")
> dischdate contains characters not specified in ignore(); no generate
>
> The same error was produced with and without the quotes
> and/or leading space in the ignore statement. Recast does
> not allow date as a type.
>
> Any thoughts on how to handle the conversion would be appreciated.
*
* 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/
