Richard Williams wrote:
>I generally find that Stata is much much smarter than I am and that I
should trust its judgment. >
Amen to that. There are a few areas where I disagree with what a procedure does or doesn't do (factor analysis being the most obvious example). But in general, to paraphrase my numerical analysis professor from grad school:
"Unless you have a damn good reason, use the library code. Surely you don't think the code you wrote in a weekend is better than what someone spent two months perfecting?" Now CS guys are stereotypically more arrogant than statisticians, who are trained in a deep disciplinary conservativism, but still the principal holds.
*
* 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/
