st: Fwd: How to manipulate crsp data to construct momentum portfolio

["Tang, Fang" <tanglion@gmail.com>]

Hi, statalist,

This question really bothers me a lot.
I want to construct 10 momentum portfolios and calculate their monthly
return.  Part of the monthly data is as follows. At the end of each
month, I have the data of price(prc), ret(current month return),
market capitalization for all the stocks--all variables possibly
missing. My methodology is as follows. At the end of month, I exclude
all stocks with 1) price< $5  2) market cap in the smallest 10%
3) no missing monthly return for previous 6 months.(Here, I understand
the sequence of these operations matters, but any choice is fine).
After this,  I split all the left stocks into 10 deciles based on
their previous 6 months cumulative return and keep track of these 10
portfolios to calculate corresponding equally weighted monthly returns
for the future 6 months. Note some of the stocks may be unlisted in
the next 6 months so missing values may emerge here. If this happens,
just reduce the number of stocks contained in each portfolio to
calculate the equally weighted monthly returns. Lastly, since all
those portfolios will be held for 6 months, at the end of each month,
I actually have 6 portfolios available in each decile.One formed right
now, one formed 1 month ago, one formed 2 months ago, and ....one
formed 5 months ago. So the real monthly return is equally weighted
return of those 6 portfolios for each decile. The final output is
monthly returns for each of the 10 decile portfolio.

I hope I made myself clear enough. Thanks very much

Stanley Tang

     +----------------------------------------------------------------------------------------------------------------------+
     |       date   year   month      cusip   ticker   permno   permco
  shrout       prc        ret        time        cap |
     |----------------------------------------------------------------------------------------------------------------------|
  1. | 01/31/1964   1964       1   04559140      AOE    29233    23606
      16    5.6875   -.031915   31jan1964         91 |
  2. | 01/31/1964   1964       1   91102630      UMO    37647    24250
      16   5.71875   -.056701   31jan1964       91.5 |
  3. | 01/31/1964   1964       1   25474130      DIL    31245    23763
      51   3.59375   -.128788   31jan1964   183.2813 |
  4. | 01/31/1964   1964       1   74099090      PMW    39001    24352
     415    .53125      .0625   31jan1964   220.4688 |
  5. | 01/31/1964   1964       1   36959520      GLE    32168    23832
      26      9.25   -.130667   31jan1964      240.5 |
     |----------------------------------------------------------------------------------------------------------------------|
  6. | 01/31/1964   1964       1   14444770      CRL    30373    23700
     318    .78125   -.074074   31jan1964   248.4375 |
  7. | 01/31/1964   1964       1   88880320      TSU    37372    24225
      97      2.75   -.083333   31jan1964     266.75 |
  8. | 01/31/1964   1964       1   88880330      TSU    37380    24225
      34    7.9375    .024194   31jan1964    269.875 |
  9. | 01/31/1964   1964       1   04999290      ASW    29364    23616
     890      .375          0   31jan1964     333.75 |
 10. | 01/31/1964   1964       1   95969010      WSI    38017    24279
     677        .5          0   31jan1964      338.5 |
     +----------------------------------------------------------------------------------------------------------------------+

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/