[R] Output interpretation: standard error of lm dummy variable
Jim Lemon
drjimlemon at gmail.com
Tue May 19 12:44:54 CEST 2015
Hi Livia,
You seem to have mixed up the residual error with the R^2, which is
just over 0.02. The bottom line on your summary table says that the
obtained F statistic was equal to 1894 (this has been truncated to
four significant places). The probability of obtaining that value with
your data given the F distribution for 4 numerator and 342199
denominator degrees of freedom is very small, but not less than 0. The
notation <2.2e-16 can be roughly translated in English as "about as
close to zero as this function can calculate". You should note that
with that many observations, a significant result is almost
guaranteed, but the linear model explains almost none of the variance
in prices.
Jim
On Tue, May 19, 2015 at 6:16 PM, Livia Maria Vestergaard
<lvest09 at student.sdu.dk> wrote:
> Hi guys
>
> I have a statistical question to an analyse I ran in R. It is a dummy variable model with the 5 regions of Denmark as 4 independent dummy variables and price as the dependent variable:
>
> price = 10.325 - 0.176*Sjaeland - 0.368 * NJylland - 0.230*MJylland - 0.120* Syddanmark
>
> I understand the R^2 = 0.7348 - that it shows the explanatory force of the model (between 0 and 1)
> My question is simply how to interpret the standard error = 0.7348 on 342199 degrees of freedom? How is it calculated when the model is a dummy variable model. And what does it mean that the F-statistic says that there are 1894 on 4 and 342199 DF (degrees if freedom?) with a p-value < 0?
>
> I have been searching for hours - and can't quite figure out how R reached the numbers and how to interpret the output of standard error and the p-value of the dummy model.
>
> I really hope you can help :)
>
> Best Livia
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
More information about the R-help
mailing list