Re: [R] ols/gls or systemfit (OLS, WLS, SUR) give identical results

From: Mihai Nica <mihainica_at_yahoo.com>
Date: Tue 18 Jul 2006 - 03:06:11 EST


  I should have not send the post to the list, and I apologize. The problem was on my part, and easily solvable. Thanks for your concern, I really appreciate it, and thanks to the list for patience. This is really a great resource and should not be abused :-)   mihai

Spencer Graves <spencer.graves@pdf.com> wrote: I can't say without more information. If the system were saturated (i.e., has as many equations as unknowns), you should get the same answer from all the different methods. However, I just tried a saturated model in 'systemfit', with the following results:

 > DF2 <- data.frame(y=1:2, x=3:4)
 > lm(y~x, DF2)
Call:
lm(formula = y ~ x, data = DF2)

Coefficients:

(Intercept)            x
          -2            1

 > library(systemfit)
 > systemfit("OLS", list(eqn=y~x), data=DF2) Error in solve.default(sigma, tol = solvetol) :  system is computationally singular: reciprocal condition number = 0  >
   If you'd like more help from this listserve, please supply a simple, self-contained example to illustrate your question (as suggested in the posting guide! "www.R-project.org/posting-guide.html").

   Hope this helps.
   Spencer Graves

Mihai Nica wrote:
> I might be sorry for asking this question :-)
>
> I have two equations and I tried to estimate
them individually with "lm" and "gls", and then in a system (using systemfit) with "OLS", "WLS" and "SUR". Quite surprisingly (for myself at least) the results are identical to the last digit.
>
> Could someone (please!) give a hint as to what
am I doing wrong?
>
> Thanks,
>
> mihai
>
> ---------------------------------
>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
                         



See the all-new, redesigned Yahoo.com. Check it out.

        [[alternative HTML version deleted]]



R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Received on Tue Jul 18 04:36:07 2006

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Tue 18 Jul 2006 - 06:16:24 EST.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.