Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Phil Clayton <philclayton@internode.on.net> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: Re: using categorical and continuous variables foreach loop for univaraite analysis |
Date | Tue, 9 Jul 2013 13:06:01 +1000 |
Like I said, you need to change your -foreach- syntax. Not: foreach x of varlist but: foreach x in Phil On 09/07/2013, at 12:51 PM, Michael Stewart <michaelstewartresearch@gmail.com> wrote: > Hi, > > I apologize, I should have told you that I have tried the solution > suggested by Phil prior to posting the question and I get an error > > ************************************************************ > foreach x of varlist i.cat1 i.cat2 con1 con2 { > logit outcome `x' > } > > factor variables not allowed > r(101); > > end of do-file > ************************************************************ > > Regarding Dimitri's solution, I have unequal number of categorical and > continuous variables.hence, will not be able to use the solution. > > > Thank you again for your time and consideration > > Mike > > On Mon, Jul 8, 2013 at 7:47 PM, Phil Clayton > <philclayton@internode.on.net> wrote: >> One simple solution is a slight tweak to your -foreach- syntax: >> >> foreach x in i.cat1 i.cat2 i.cat3 i.cat4 con1 con2 con3 con4 { >> logit outcome `x' >> } >> >> Phil >> >> On 09/07/2013, at 8:51 AM, Michael Stewart <michaelstewartresearch@gmail.com> wrote: >> >>> Hello all, >>> >>> In my research project, I have a number of categorical predictors(cat1 >>> cat2 cat3 cat4 ) and continuous predictors(con1 con2 con3 con4) and >>> during initial uni-variate analysis, I am using two step process to >>> evaluate for statistical significance.My code is a follows >>> >>> I use stata 12 >>> >>> *********************MY CODE ******************** >>> foreach x of varlist cat1 cat2 cat3 cat4{ >>> logit outcome i.`x' >>> } >>> >>> foreach x of varlist con1 con2 con3 con4{ >>> logit outcome `x' >>> } >>> ******************************************************** >>> Question: >>> >>> Is it possible to achieve this is one step i.e., use continuous and >>> predictors as a single varlist but perform logistic regression still >>> respecting the fact that some are categorical variables and some are >>> continuous variables >>> >>> -- >>> Thank you in advance for your time and consideration, >>> Yours Sincerely, >>> Mike. >>> * >>> * For searches and help try: >>> * http://www.stata.com/help.cgi?search >>> * http://www.stata.com/support/faqs/resources/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/faqs/resources/statalist-faq/ >> * http://www.ats.ucla.edu/stat/stata/ > > > > -- > Thank you , > Yours Sincerely, > Mike. > * > * For searches and help try: > * http://www.stata.com/help.cgi?search > * http://www.stata.com/support/faqs/resources/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/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/