R code: Error in parse(text = x, keep.source = FALSE) : :2:0: unexpected end of input

Question

I would like to run a multi logit model in R. The data has been transformed using the mlogit.data() function.

dt1 = mlogit.data(dt, choice = "Buy", shape = "long", chid.var = "Id", 
                 alt.levels = c("model1","model2","model3","model4"))

But when I later run the multi logit model using mlogit()

mlogit(Buy~norRebate,data=dt1)

An error shows up as:

Error in parse(text = x, keep.source = FALSE) : 
  <text>:2:0: unexpected end of input
1: . ~   . |  
   ^

But the dt1 is okay if I run it using gmnl(). What's wrong with the code or data format when I run it using mlogit() and what's the difference running multi logit model between mlogit() and gmnl()?