Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | Fernando Rios Avila <f.rios.a@gmail.com> |
| To | statalist@hsphsun2.harvard.edu |
| Subject | Re: st: Question about the use of ratio variables |
| Date | Mon, 9 Jul 2012 22:43:43 -0400 |
I think your problem is similar to the division bias issue in Labor economics. I suggest you to look for the following paper: The Relationship between Wages and Weekly Hours of Work: The Role of Division Bias George J. Borjas The Journal of Human Resources Vol. 15, No. 3 (Summer, 1980), pp. 409-423 It discusses the problem, consequences and some solutions. Fernando. On Mon, Jul 9, 2012 at 10:20 PM, hsini <hsini92@gmail.com> wrote: > > Dear all, > > I have a question about the use of ratio variables. > Is it problematic to have a regression equation like : (Yi/population) = b0+ b1*(Xi/Yi) +b2X2 + error? > > Let's say Yi is the number of innovation per region, and Xi is the number of small firm innovation. The ratio variable is a very important measure since in my theory it represents the organizational ecology of the region. I excluded regions with zero innovation. However, I am afraid of any function problem since I have Yi on the both side of the equation. I have read Bradshaw and Radbill (1987) paper discussing the use of ratio variables. It seems to me they were dealing with the ratio is generated by a third variable (Z), which is a bit different from my situation. Any suggestion? > > > Sincerely, > Hsini Huang > * > * 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/