search

pbreheny / visreg

Visualization of regression functions
http://pbreheny.github.io/visreg/
61 stars 18 forks source link

error in parse #102

Closed nath2801 closed 1 year ago

nath2801 commented 2 years ago

Hi,

I have a little problem! I want to use the dataset "freMTPL2fre" from the package CASdatasets and I want to run this code...

m1 <- glm(ClaimNb ~ (offset(log(Exposure)))+ VehGas + VehBrand , data = freMTPLfre,family=poisson(link = log)) visreg(m1,"VehGas", type="contrast", scale="response", gg=TRUE, rug=FALSE, partial=FALSE) + theme_bw()

I have an error:

Erreur dans parse(text = c("cond=c(cond, list(", names(f)[j], "=factor(mode, levels=levels(f[, names(f)[j]]))))")) :

:3:1: '=' inattendu(e) 2: NA 3: = But for example, if I take the dataset "freMTPLfre" from the same package and I want run the same code... m1 <- glm(ClaimNb ~ (offset(log(Exposure)))+ Gas + Brand , data = freMTPLfreq,family=poisson(link = log)) visreg(m1,"Brand", type="contrast", scale="response", gg=TRUE, rug=FALSE, partial=FALSE) + theme_bw() I don't have any problem Can someone help me?
pbreheny commented 2 years ago

I don't understand what you're asking. Gas and Brand are variables in the data set "freMTPL2freq". VehGas and VehBrand are not. You can't fit models on variables that don't exist, let alone plot them.

If there are extra steps you're not showing where these variables are created, no one can help you unless you supply a reproducible example.

pbreheny commented 1 year ago

Closing this issue since I never heard back from you.