I want predict a fitted values of a probit model for a data panel. My model is:
m.random<- pglm(formula = pooled,
index = c("code","Year"),
data = data_modelo,
family = binomial('probit'),
model = "ramdon",
method = "bfgs",R = 5)
The predict
command doesn't help me, it says there is no applicable method to 'predict' applied to an object of class "c ('maxLik', 'maxim')"
. The margins
command also does not estimate the marginal probabilities. Can someone help me with command or techniques to predict the values and estimate the marginal probability?