OLS regression with heteroscedasticy problem
by 3J LEMA
For my panel data (all continuous variables; 1 DV, 4 IVs; 8 time points for
each of the 5 companies), the Breusch-Pagan test is nonsignificant
(Chi-square(4) = 1.77291, p= 0.777434) implying that both Random effects
model and fixed effect model are not appropriate. I used the OLS
regression with heteroscedasticy problem. Any suggestions on how to remedy
this?
Appreciate your help.
Lema.
2 years
Unprofessional people in this mailing list
by 3J LEMA
People in this group are mostly too arrogant, egoistic, unprofessional,
etc. These people feel that they are brilliant, wise, intelligent, smart,
and knowledgeable. I doubt if they are really knowledgeable in
the field. Don't ask me why; you can identify them by reading their
comments. Why give comments when your words are not helpful to
other people?
2 years
v0.3 of PanelTools out now
by Artur T.
Here is another announcement:
Version 0.3 of the PanelTools package is available now:
http://gretl.sourceforge.net/current_fnfiles/PanelTools.gfn
# Changelog
v0.3 (03-2022)
- Add new functions mpmean(), meanc_robust() and sdc_robust()
- Add new function fe_estimator() for estimating a fixed-effects model
using matrices as inputs (no inference supported, yet)
- Add new function pols_estimator() for estimating a pooled-OLS model
using matrices as inputs
- Add new function pfcast() for computing predictions using some
supported panel model using matrices.
Please report bugs or feature requests on the mailing list or the github
repo: https://github.com/atecon/PanelTools
Best,
Artur
2 years
