Hi,
I am trying to carry out a multinomial regression using the cumlogit link
function. I have tried using the VGAM package, and have gotten some results...
fit1 <- vgam(Y ~ X1 + X2 + X3 + X4,
cumulative(link=logit,intercept.apply=FALSE,parallel=TRUE),
data = data1
)
The problem arrises when I try to get the information out of the fitted object.
I can get the coefficient estimates using:
coef(fit1)
When using the summary, I get the following table:
Number of linear predictors: 5
Names of linear predictors: logit(P[Y<=1]), logit(P[Y<=2]),
logit(P[Y<=3]), logit(P[Y<=4]), logit(P[Y<=5])
Dispersion Parameter for cumulative family: 1
Residual Deviance: 618.6155 on 1671 degrees of freedom
Log-likelihood: -309.3078 on 1671 degrees of freedom
Number of Iterations: 11
DF for Terms and Approximate Chi-squares for Nonparametric Effects
Df Npar Df Npar Chisq P(Chi)
(Intercept):1 1
(Intercept):2 1
(Intercept):3 1
(Intercept):4 1
(Intercept):5 1
X1 1
X2 1
X3 1
X4 1
My question is, how do I get the P values ? I have looked through the structure
using str(fit1), but I seem to be unable to find it. I can't find the
standard error of the estimates either. Am I missing something ?
Thanks.
Robert
_________________________________________________________________
[[alternative HTML version deleted]]
