Re: st: Collapsing daily data to a monthly data on the 15th each month

[Nick Cox <njcoxstata@gmail.com>]

Let's identify each month by a start date that is 16th of each month
and each year.

The target starts are

gen target = date if day(date) == 16

Some surgery will needed at the beginning to enter an appropriate
target for observation 1 if the first day is not the 16th.

replace target = mdy(<month>, 16, <year>) in 1

Then we copy each target by

replace target = target[_n-1] if missing(target)

then -collapse ... , by(target)-

Nick
njcoxstata@gmail.com


On 21 January 2014 17:42, Toby Knapen wrote:
> Dear Statalisters,
>
> I would like to transform my daily data to monthly data, calculated on
> the 15th of each month, more precisely, from the 16th day of a
> specific month up to and including the 15th day of the following
> month. The problem is that I can only find the way to this it based on
> 'regular' months. This can be done by using the following commands:
>
> g dm=mofd(date)
> format dm %tm
> collapse  prc ret shrout vwretd ewretd, by (dm)
> tsset dm, monthly
>
> Since I'm having inflation dates on the 15th each month I would like
> to have the monthly data on the 15th as well, otherwise there will be
> a mismatch in my data. The data on my dates looks like below:
>
> 16dec2002
> 17dec2002
> 18dec2002
> 19dec2002
> 20dec2002
> 23dec2002
> 24dec2002
> 26dec2002
> 27dec2002
> 30dec2002
> 31dec2002
> 02jan2003
> 03jan2003
> 06jan2003
> 07jan2003
> 08jan2003
> 09jan2003
> 10jan2003
> 13jan2003
> 14jan2003
> 15jan2003
> 16jan2003
> 17jan2003
> 21jan2003
> 22jan2003
> 23jan2003
> 24jan2003
> 27jan2003
> 28jan2003
>
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/