I have made a logistic model for the probability a child will be immunized.
Here is the model:
M3 <- lmer(immun ~ kid2p + order23 + order46 + order7p + indNoSpa + indSpa +
momWork + rural + pcInd81 + (1|mom), family=binomial(link="logit"))
'mom' designates a child's family, where all the families have been indexed
I'm trying to predict the probability a new born child to a family will be
immunized. Here's what I have:
new.info<- data.frame(kid2p=0, order23=0, order46=1, order7p=0, indNospa=0,
indSpa=0, momWork=1, rural=0, pcInd81=.0086511, mom=245)
pred.interval<-predict(M3, new.info, level=.95)
This function supposedly gives me the 95% confidence interval and
probability estimate. I could just take the regression coefficients and
compute the probability manually, but I'm not sure how to deal with the
'mom' input varying by intercept. Can I pull out the intercept value for
mom=245? If so, I could just compute it all based on the regression summary
and a simulation.
-chris
[[alternative HTML version deleted]]
______________________________________________
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
