Dear list
I have a data set involving binary responses (successes failures) for which
some explanatory variables result in a quasi complete separation problem.
To deal with the separation problem I tried to run a glm with
"bayesglm" in
the arm package.
However when I try to compare different bayesglm models, the anova doesn't
give any value.
> anova(model1,model2,test="Chi")
Analysis of Deviance Table
...
Resid. Df Resid. Dev Df Deviance Pr(>Chi)
1 336 118.67
2 336 118.68 0 -0.0069818
Does anyone know how to solve this problem ?
Thank you in advance for your time and for any advice you could provide
Best regard
--
View this message in context:
http://r.789695.n4.nabble.com/Model-comparison-with-bayesglm-tp4637989.html
Sent from the R help mailing list archive at Nabble.com.