Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Tirone, Vanessa" <vtirone@utk.edu> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | st: dropped time variable |
Date | Tue, 9 Apr 2013 01:47:06 +0000 |
Hello, I am attempting to run a model with one independent variable and several control variables predicting a forward lagged variable. There is data on the predictors and outcome variables at baseline, 6 months, and 12 months. I was also planning on running another model containing an interaction term between the independent variable and time. However, in the first model the time variable is dropped from the model due to collinearity. Is there a way to resolve this? Does it say something about the stability of my variables over time? Below is a copy of my syntax. The outcome variable is zcom. reshape long Zcom Zalt Zinv Zsat Zphys Zpsyc Zsex Zinj, i(Participant) j(time) tset Participant time *forward lagged variables g flZcom = f.Zcom g flZalt = f.Zalt g flZinv = f.Zinv g flZsat = f.Zsat g flZphys = f.Zphys g flZpsyc = f.Zpsyc g flZsex = f.Zsex g flZinj = f.Zinj gen inv_x_time = Zinv*time xtreg flZcom Zcom Zinv time Zphys Zpsyc Zsex Zinj, pa corr(ind) robust xtreg flZcom Zcom Zinv time inv_x_time Zsat Zalt Zphys Zpsyc Zsex Zinj, pa corr(ind) robust Thank you, Vanessa Tirone, M.A. University of Tennessee Clinical Psychology Doctoral Candidate * * 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/