Exactly what I expected, if CLRM means what I guess it means.
But if so, why do you expect the F test to still work?
Niels Bohr had a story of a physicist who had a horseshoe
over his door for good luck, but said -- when reproached
for superstitious nonsense -- "I've been told
that it works even if you don't believe in it."
Is that your position?
Nick
[email protected]
M. Haider Hussain
Errors were non-normally distributed, that's why I'm using -qreg-. In
other words, I'm not prepared to enforce CLRM assumptions.
Haider
Social Policy and Development Center
Karachi, Pakistan.
> What model is this based on? Gaussian/normal
> errors? If you are prepared to buy that, why
> are you doing -qreg-?
>
> Nick
> [email protected]
>
> M. Haider Hussain
>
> > After running bootstrapped quantile regression with k=15, n=5401, I
> > obtained Pseudo R2=0.3161. If I want to compute the joint significance
> > of the regressors, can I still use the F-test given by
> >
> > F=[r2/(k-1)]/[(1-r2)/(n-k)]
> >
> > If this isn't the case, what's the measure of joint significance of
> > the regressors after -qreg- / -bsqreg-?
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
