Stephen P. Jenkins wrote
> I am going through the code for -sureg-
> in detail (hoping to adapt some of it
> for my own uses!). (Similar code appears
> in related commands.) I have two queries:
> (1) Why is there no -markout- line to
> accompany the -markout- one?
> [Is it because this is handled via
> -reg3-, which -sureg- calls?]
I am not sure what you mean by "accompany the -markout- one" since -markout-
does not appear in -sureg-. Did you mean to write "accompany the
-marksample touse- line"? If so, then note that -marksample touse- is
equivalent to typing
tempvar touse
mark `touse' `if' `in'
markout `touse' `varlist'
Therefore, -markout ...- is not required if the aim is to define the binary
variable `touse' as being ==1 when the [if exp] and [in range] conditions
are satisfied and the variables in `varlist' are not-missing, ==0 otherwise.
That said, nothing prevents the programmer to use -markout `touse'- after
-marksample- when one also wants to restrict the sample to observations
where values of other variables (ie. variables not in the varlist parsed by
-syntax ...-) are not missing. See [P] mark or -help mark-.
Patrick Joly
joly.patrick@ic.gc.ca
pat.joly@utoronto.ca
*
* 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/