Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | daniel klein <klein.daniel.81@googlemail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: probing interactions with continuous variables with mi test / mi test transform |
Date | Fri, 10 Jun 2011 09:53:05 +0200 |
Without going into the statistical issues, please read the error messages and try to make sense out of them, which is usually not too hard. If Stata tells you, it cannot find a variable "mi" you should check why Stata is interpreting any "mi" in your syntax as a variable. I guess it is because you used too many "////". Your code says: [...] trustmleC knowlmleC knowlmleCXbsent3ae /// mi test sdb avg sda /// Since the command (mi estimate : regress) does not end after the first line, Stata searches for the next variable in the secound line, which is "mi" and which does not exist. In Stata "///" means: the command is not yet finished. So you should get rid of "///" in _both_ lines. In your last example your open brace is "[", where Stata expects "(" as used in Sytaxes above. Stata looks for the mi subcommand "[diff1", does not find it, and tells you about it. Best Daniel * * 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/