Issue with rstanarm stan_lm


#1

Hello,

I am trying to fit a simple stan_lm on some simulated data and I’m getting an error that says “please report bug”. Any ideas about what is going wrong?

library(rstanarm)

X = matrix(rnorm(500), nrow = 100)
Y = 3+X%*%c(1,2,3,0.5,0.5)

data = as.data.frame(cbind(Y,X))

model <- stan_lm(V1~V2 + V3+ V4+ V5+V6, data = data, prior = R2(location = 0.2))
model


#2

Your model has no error term and thus it cannot initialize with the default values, which is what all the preceding warning messages are about. I would recommend trying

Y <- 3 + X %*% c(1,2,3,0.5,0.5) + rnorm(500)

or something similar.