[Gretl-users] Covariance Matrix with standard errors, and finding the p-value of the F-statistic when taking out regressors

[Allin Cottrell]

On Sun, 30 Mar 2008, Ben wrote:

> I'm an undergrad, just started looking at gretl. It looks 
> amazing; thanks a bunch for the work.

Thanks, glad you like it!

> I have two minor suggestions.
> 
> 1) I noticed you can make a covariance coefficient matrix. 

A coefficient covariance matrix?  This is a matrix showing the 
estimated variances and covariances of the coefficient estimates.

> Have you considered including the standard errors under these, 
> and also using the asterisks to mark whether the coefficients 
> are statistically significant at some level?

Standard errors and "significance stars" for the regression 
coefficients are shown as a regular part of regression output.  
I'm not sure what you're asking for here.

> That's how Stock and Watson seem to present these coefficient 
> matrices, and they also include summary statistics at the bottom 
> (SER, R-bar squared, and n).

Ah, maybe you're thinking of something like the gretl "model 
table" (rather than a covariance matrix).  Try checking that out.
It's described in section 3.4 of the User's Guide.

> 2) The output of a least squares regression is somewhat long, 
> and I've noticed that I can't selectively copy/paste from it. 

I think you can copy a selected chunk with Ctrl-C.  I'll check 
that the next time I'm in Windows.

> I've noticed that in general I can't 
> right-click on these outputs, and I wonder why.

There's not much to offer on right-click, but I guess we could add 
a Copy option there.

> 3) It would also be nice to just output an equation that I could 
> copy/paste, or export somehow.

That's supported if you have LaTeX installed: Model window 
/LaTeX/Copy/Equation.  

> I imagine this is the best open-source stats program around? 

Yes, of course ;-)  Well, R is very good in its own way
http://cran.r-project.org

> Can gretl do time-series well?

Yes.  Its not totally comprehensive but has pretty full 
time-series support.  Take a look at Chapter 20 in the User's 
Guide, for example.

> How does gretl compare to the commercial software (I've used 
> Eviews 4.1 Student Version and SPSS, and both seem terribly 
> clunky).

There are some things Eviews does that gretl doesn't (and the same 
for SPSS).  (There are also some things gretl does that the others 
don't.)  Gretl's numerical accuracy has been thoroughly vetted by 
Talha Yalta.  As for the interface, I agree with your comment on 
clunkiness.

Allin Cottrell