predict probability in r

[sharonqu]

I want to predict the probability in glm ,but the console reported that unexpected symbol in AA which could not be find. How can I solve it?
this is my code.

model2 <- glm(survived ~age+sibsp+parch+pclass+embarked+fare+sex,family=binomial(link='logit'),data=mdata)
summary(model2)

head(predict(model2, type = "link"))
#predict probability
head(predict(model2, type = "response"))
AA <- data.frame(sibsp = 2,age =21,parch =2,pclass = 1,embarked ="c", fare = 26.55 sex ="female"))
predict(model2,type = "response",newdata = AA)
predict

[joels]

The code to create AA is missing a comma after 26.55. So it should be:

AA <- data.frame(sibsp = 2,age =21,parch =2,pclass = 1,embarked ="c", 
                 fare = 26.55, sex ="female"))

[sharonqu]

But now I have a new question. After i fixed it ,the console reported this for me.

Error in model.frame.default(Terms, newdata, na.action = na.action, xlev = object$xlevels) :
factor embarked has new level c

How can i solve it?

[nirgrahamuk]

You should understand the factor levels of your data, so that when producing new data to predict on that they are factors with the same levels.

levels(mdata$embarked)
levels(mdata$sex)

Use the levels command to find out the levels that were present when the model was built