Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | John Luke Gallup <jlgallup@pdx.edu> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: Outreg margins error |
Date | Thu, 6 Sep 2012 19:17:53 -0700 |
Wameq, If you were using a nonlinear regression method instead of -regress-, I would think you would want to use the following -margins- command since you don't seem interested in factor variable interactions: margins, dydx(*) As it is, in OLS the marginal effects are identical to the coefficients. By the way, -outreg- is now updated on SSC resolving the bug that you identified. John On Sep 6, 2012, at 8:20 AM, Nick Cox <njcoxstata@gmail.com> wrote: > I don't think you are understanding what -margins- does. Look at the > help and you will see that it starts > > margins [marginlist] [if] [in] [weight] [, response_options options] > > where marginlist is a list of factor variables or interactions that appear > in the current estimation results. > > So, you are offering > > tfhhh shg lntotalexp ... > > to -margins- as a marginslist. My interpretation is that -margins- is > happy to accept -tfhhh shg-, whatever they are, as part of a > marginslist but bails out because -lntotalexp- is certainly not > acceptable. > > Otherwise put, it's not enough that your syntax looks like what > Carsten was using as an example. It must also make sense to -margins. > Also, you really must read the help! > > Nick > > On Thu, Sep 6, 2012 at 4:08 PM, Wameq Raza <wameq.r@gmail.com> wrote: >> Dear Nick, >> Thanks for the response. Yes, the lntotalexp is the natural log of >> expenditure. My syntax was: >> >> margins tfhhh shg lntotalexp lnhhsize fem0to13 fem36to55 fem55 >> men0to13 men14to35 men36to55 men55 eduprim_hh edumid /// >> _hh eduhigh_hh scst dumemp_hh2 dumemp_hh3 dumemp_hh4 dumemp_hh5 >> dumemp_hh6 kanpur allahabad aillness1 aillness2 ail /// >> lness4 aillness5, post >> >> lntotalexp: factor variables may not contain noninteger values >> r(452); >> >> >> Hopefully this syntax is correct following the previous response from >> Carsten. I've also tried getting the margins using the same syntax >> one variable at a time, but to no avail. Any suggestions? >> Wameq >> >> On Thu, Sep 6, 2012 at 2:02 PM, Nick Cox <njcoxstata@gmail.com> wrote: >>> My suggestions are: >>> >>> 1. The error means what it says: -lntotalexp- (presumably the log of >>> total expenditure) cannot be a factor variable. >>> >>> 2. The error arises from incorrect syntax, but you need to show us >>> _exactly_ what you typed with your data. >>> >>> 3. For "STATA" read "Stata" throughout. >>> >>> 4. Points 2 and 3 are covered by the Statalist FAQ, so you really >>> should read that before re-posting. >>> >>> Nick >>> >>> On Thu, Sep 6, 2012 at 11:01 AM, Wameq Raza <wameq.r@gmail.com> wrote: >>>> Dear Carsten and John, >>>> Thank you so very much for the replies. As for the syntax Carsten >>>> suggested, STATA gives me an error when estimating continuous >>>> variables such as the ln of expenditures. The error message it gives >>>> me is: >>>> >>>> -lntotalexp: factor variables may not contain noninteger values- >>>> >>>> Any suggestions for this? >>>> >>> * >>> * 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/ >> >> >> >> -- >> W A M E Q R A Z A >> * >> * 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/ > * > * 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/ * * 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/