Dear colleagues,
thank you very much for your answers.
My variables are dummy coded so my model is not affected by the
"problem" of linear combination (I guess so).
Joel: do I get you right that the collinearity can occur if the
distribution of some of the variables is very unbalanced and one value
prevails?
Thank you very much for your help!
Joan
Joel J. Adamson writes:
Di Stefano, Paola writes:
Dear Joan,
Did you check if one variables is linear combination of another? It
could happen when a variable is created using the others.....
Paola
Good thinking: as I said the linear combination that comes up for me
most often is 0.
Joel
Joan Holand writes:
Dear statalist,
I'm running a loglinear model using the -glm- command (Stata 9SE,
Windows XP):
glm depvar a*b*c*d, fam(pois)
I'm specifying 4-way-interactions but unfortunately some of the
coefficients are beeing dropped because of the collinearity. I'm
sure that I did not use the same variables twice.
Assuming that my model is ok. :-) , what could be other possible
reasons for dropping coefficients?
Thank you very much for any comments! Joan * * 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/
*
* 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/
