[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: RE: Taking averages, etc.
At 12:38 PM 12/17/2003 +0000, Allan Reese wrote:
I'm not sure how that performs fewer tests. The if-then-else structure
only performs as many tests as are necessary whereas the above would
perform 2 tests for each case. In any event, I think I'd be more likely to
make errors with syntax like this. But that may just be me!
First point is to analyse the logic. Reforming a problem often speeds up
the execution far more than fancy coding, and leads to insights. In this
replace y=3 if x1==1 & x2==3
replace y=4 if y==5 & (x3==2 & x4==17)
performs far fewer tests. The parentheses do not affect the result but
emphasise the logic.
You must get Nick's messages faster in the UK! I look forward to what he
has to say. Thanks Allan.
Stata, as Nick has already replied, has a completely general control
language that includes if/then/else, for and while. Unlike SPSS, where
only data transformations can be put into DO IF and LOOP constructions,
Stata allows any statements or blocks of statements. These may
conveniently be written using an editor (provided it doesn't add a file
extension!) and saved as .DO or .ADO on the fly.
Richard Williams, Associate Professor
OFFICE: (574)631-6668, (574)631-6463
WWW (personal): http://www.nd.edu/~rwilliam
WWW (department): http://www.nd.edu/~soc
* For searches and help try: