Re: st: Generating new variable by sorting and taking the difference of extreme observations every month

[Sergiy Radyakin <serjradyakin@gmail.com>]

David,

1. what is the role of "quarterly financial statements" in this assignment?
2. I understand that "market data"="security"="total assets"=v, where
v is a [one, single] numeric variable you have, which varies by
country, period, firm_id. True of false?
3. from what you describe your task is separable by countries and
months. if this is not true, explain the dependency.
4. reduce the complexity first by writing the program for the case of
one country and one month, then make the program 'byable' (see help)
and then add a loop to run it by(country month).
5. My understanding is that returns task is the next step after the
classification into top and bottom groups is done. So do the
classification first. Are the returns quarterly or monthly?

Best, Sergiy Radyakin



On Sat, Nov 9, 2013 at 4:32 AM, David Ashcraft <ashcraftd@rocketmail.com> wrote:
> Hello Statalist,
>
> I have quarterly financial statement and monthly market data for about 1800 firms from six countries for the period 2000-2013. Now I want to create a variable that will sort every month all securities for each of the six countries by total assets. Once sorted take the upper and lower ten percent of the companies and generate a new variable by subtracting the returns of top ten percent from the lower ten percent as sorted by total assets. Any help will be highly appreciated.
> Best Regards
>
> David
> --
> David Ashcraft
> PhD Candidate
> Bangor University
>
> *
> *   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/