Re: st: The philosophy behind storing missing values as very large positive numbers

[Richard Williams <Richard.A.Williams.5@nd.edu>]

At 03:21 AM 7/3/2004 -0400, jockmcdock@netscape.net wrote:
> One thing in Stata that irks me is the treatment of missing values. Stata 
> (IMO) does not handle missings consistenly. For example, if I issue a 
> regression command, missings are recognised as missings and these values 
> are not included in the regression. However, if I type
>
> gen x=1 if (y>10)
>
> then x has the value 1 even if y is missing. In other words, Stata treats 
> as a real value. It's a problem that's easy to solve, but if you forget to 
> include "& y!=.", you can create problems further down the line. Other 
> statistical packages get this right.
As a sidelight, with the new extended MD codes "& y!=." is potentially 
problematic coding, as it will not exclude .a, .b, etc.  Better is

gen x=1 if (y>10) & !missing(y)

or else

gen x=1 if (y>10) & (y < .)

And as another sidelight, as far as other packages go, SPSS gets it "right" 
(as you define it) but I think it was established in some other thread that 
SAS gets it "wrong".  But, not being a SAS user, I can't confirm that.


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc

*
*   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/