Re: st: Creating Standard Deviation variable for daily panel data (stock returns)

["C. Evans" <ce306@cam.ac.uk>]

Dear Nick

I would like standard deviation values of returns calculated for each id 
each month. So I would have a value of 0.2 everyday in one month for id=1 
and then the 2nd month a different standard deviation. I should see a 
different SD each month and for each different Id - probably.

I'm not very good at explaining

Thanks,
Chris

On Jul 3 2013, Nick Cox wrote:

> I missed in this a statement of what you _do_ want. If you have a
> block of values for which you calculate the SD, it's necessarily the
> same result for all observations contributing.
>
> But  Stata won't take account of missing values it does not have.
>
> Nick
> njcoxstata@gmail.com
>
> On 3 July 2013 17:37, C. Evans <ce306@cam.ac.uk> wrote:
>
> > I've been using the command egen sdreturn3 = sd(return), by(date) to 
> > try and create a new variable that takes the standard deviation of 
> > return for that month or even that day. This gives me the same sd for 
> > every id at that date (which I don't want)
> >
> > my panel looks like this
> >
> > id    date     returns
> > 1   01Jan1964    0.1
> > 1   02Jan1964    0.12
> > 1   03Jan1964    0.05
> > 2   01Jan1964    0.3
> > 2   02Jan1964    0.25 2   03Jan1964    0.1
> > 3   01Jan1964    -0.1
> > 3   02Jan1964    -0.2
> > 3   03Jan1964    -0.3
> >
> > Hopefully the format has turned out OK. The values are just there for an
> > example.
> >
> > Also, does it matter when I use xtset, id date, daily that I am using
> > business data and thereore the weekends are not included in my dataset
> *
> *   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/
*
*   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/