bic.glm predict error: "newdata is missing variables" - r

Question

I've spent a lot of time trying to solve this error and searching for solutions without any luck, and I thank you in advance for your help.
I'm trying to create predicted values from the coefficients created via BMA. Whenever I run my predict function, I am getting a "newdata is missing variables" error. All variables included in the original model are present in the new dataframe, so I'm not quite sure what the problem is. I'm working with a fairly large dataset with many independent variables. I'm fairly new to R, so I apologize if this is an obvious question!
y<-df$y
x<-df
x$y<-NULL
bic.glm<-bic.glm(x, y, strict=FALSE, OR=20, glm.family="binomial", factortype=TRUE)
predict(bic.glm.bwt, x)
I've also tried it this way:
bic.glm<-bic.glm(y~., data=df, strict=FALSE, OR=20, glm.family="binomial", factortype=TRUE)
predict(bic.glm, x)
And also with creating a new data frame...
bic.glm<-bic.glm(y~., data=df, strict=FALSE, OR=20, glm.family="binomial", factortype=TRUE)
newdata<-x
predict(bic.glm, newdata=x)
Each time I receive the same error message:
Error in predict.bic.glm(bic.glm, newdata=x) :
newdata is missing variables
Any help is very much appreciated!

Answer 1

First, it is bad practice to call your LHS the same name as the function call. You may be masking the function bic.glm from further use.
That minor comment aside... I just encountered the same error. After some digging, it seems that predict.bic.glm checks the names vs. the mle matrix in the bic.glm object. The problem is that somewhere in bic.glm, if factors are used, those names get a '.x' or just '.' appended at the end. Therefore, whenever you use factors you will get this error.
I communicated this to package maintainers. Meanwhile, you can work around the bug by renaming the column names of the mle object, like this (using your example):
fittedBMA<-bic.glm(y~., data=df)
colnames(fittedBMA$mle)=colnames(model.matrix(y~., data=df)) ### this is the workaround
predict(fittedBMA,newdata=x) ### should work now, if x has the same variables as df

Answer 2

Okay, so first look at the usage section in the cran documentation for BMA::bic.glm.
here
This example is instructive for a data.frame.
Example 2 (binomial)
library(MASS)
data(birthwt)
y <- birthwt$lo
x <- data.frame(birthwt[,-1])
x$race <- as.factor(x$race)
x$ht <- (x$ht>=1)+0
x <- x[,-9]
x$smoke <- as.factor(x$smoke)
x$ptl <- as.factor(x$ptl)
x$ht <- as.factor(x$ht)
x$ui <- as.factor(x$ui)
bic.glm.bwT <- bic.glm(x, y, strict = FALSE, OR = 20,
glm.family="binomial",
factor.type=TRUE)
predict( bic.glm.bwT, newdata = x)
bic.glm.bwF <- bic.glm(x, y, strict = FALSE, OR = 20,
glm.family="binomial",
factor.type=FALSE)
predict( bic.glm.bwF, newdata = x)