Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Nick Cox <njcoxstata@gmail.com> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | Re: st: Sorting data in deciles and then regressing and storing coefficients. (Looping) |
Date | Wed, 10 Jul 2013 14:33:46 +0100 |
Chris Evans: You often, indeed usually, employ the incorrect name "Satalist" in postings. Why on earth? If the reason is only that you don't check your posts carefully, it doesn't encourage careful attention to your questions. The -egen- function -xtile()- is user-written and to be found in the package -egenmore- (SSC). The Statalist FAQ explains that you should tell us where user-written programs you use come from. Yet further advice you are ignoring is to edit posts so that only the important part remains. That aside, the command bysort year size: egen portfolio = xtile(size), nq(10) makes no sense. Within groups defined by distinct values of -size-, -size- is necessarily constant and there is no point to a quantile-based subdivision. You might as well say: give me all the deciles of 42. Answer: they are all 42. I don't know what -sample()- refers to. Nick njcoxstata@gmail.com On 10 July 2013 13:59, C. Evans <ce306@cam.ac.uk> wrote: > Dear Satalist > > Yes I think you understand it correctly. The problems you have identified > are the problems I was having conceptually. Yet stata command wise I was > even struggling with creating these portfolios (remembering I am trying to > create new portfolios every year dependent on the variable size). The code I > tried to use was: > > bysort year size: egen portfolio = xtile(size), nq(10) > > this brings up the error value "too many variables" Is there a way around > this? I was thinking of -sample()- do you know if this is the only option I > have at this point to create the portfolios? > > > Thanks for your consideration, > Chris > > On Jul 9 2013, Jeph Herrin wrote: > >> So if I understand this model >> >> R = constant + beta*RM >> >> it is estimated using all the stocks in the portfolio? in which case it >> doesn't make sense that R is constant for that portfolio, as you suggest. On >> the other hand, if the model is estimated using the 10 portfolios as >> observations, it's not clear how RM is constructed for each portfolio. * * 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/