st: missing data, tab ..., matcell(...) and marksample

["Paul A. Jargowsky" <paul.jargowsky@utdallas.edu>]

I have written a program to replace tabulate, but adds the ability to 
format the output, whether frequencies or row/col/cell percents.  (Oddly, 
there isn't a way to do this in Stata.  tabulate doesn't allow formatting 
of output, and table doesn't allow row, col, or cell percents).  The 
program also organizes the output a little differently, more to my liking.

The basic operation of the program is that "tab ..., matcell(X)" is invoked 
internally to produce a matrix of frequencies; these are then manipulated 
with matrix operations.

The problem I am having concerns missing values in the row or column 
variable.  I won't to be able to retain them, but when I specify the 
"missing" option to my table command, the missing observations are ignored. 
I have discovered that the problem is the marksample command, which is 
needed for byableness and ifableness.  In other words, in the following 
code:

marksample touse
....
.....
......
quietly tab `1' `2' [`weight' `exp'] if `touse',
	matcell(`X') matrow(`R') matcol(`C') `missing'
.......

the output matrix `X' will not have a row for the missing category of the 
row or column variable even if there should be one.  But if I comment out 
"if `touse'", it works:

quietly tab `1' `2' [`weight' `exp'] /* if `touse'*/,
	matcell(`X') matrow(`R') matcol(`C') `missing'

Here is the output from the two versions of the do file, with a rigged data 
set:

With "if `touse'":
. pjt row col, col miss lines

-----------------------------------------------------------
          *** row by col (Column Percents) ***
-----------------------------------------------------------
                               col
row                 1           2           3       Total
-----------------------------------------------------------
  1              33.3        33.3        33.3        33.3
  2              33.3        33.3        33.3        33.3
  3              33.3        33.3        33.3        33.3
-----------------------------------------------------------
  Total         100.0       100.0       100.0       100.0
-----------------------------------------------------------

Without:
. pjt row col, col lines miss

-----------------------------------------------------------------------
                *** row by col (Column Percents) ***
-----------------------------------------------------------------------
                                     col
row                 1           2           3           .       Total
-----------------------------------------------------------------------
  1              25.0        25.0        25.0        33.3        26.7
  2              25.0        25.0        25.0        33.3        26.7
  3              25.0        25.0        25.0        33.3        26.7
  .              25.0        25.0        25.0         0.0        20.0
-----------------------------------------------------------------------
  Total         100.0       100.0       100.0       100.0       100.0
-----------------------------------------------------------------------

Is this a feature or a bug?

Paul Jargowsky



=========================================================================
Paul A. Jargowsky, Ph.D., Assoc. Prof. of Political Economy
Director, The Bruton Center, School of Social Sciences (GR 31)
University of Texas at Dallas, 2601 North Floyd Road, Richardson TX 75080
=========================================================================
email: jargo@utdallas.edu or Paul.Jargowsky@utdallas.edu
Home page: http://www.utdallas.edu/~jargo
Voice: 972-883-2992; FAX: 972-883-2735
=========================================================================
*
*   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/