I don't see an obvious reason for this so please try to post this as an issue on GLM.jl.
2015-02-27 10:47 GMT-05:00 Andrew Newman <[email protected]>: > Hi Julia-users, > > I am trying to run a few simple regressions on simulated data. I had no > problem with a logit and was able to run it using glm and subsequently > match the results with a separate maximum likelihood. I'm having a lot > more trouble with the probit which doesn't seem to want to converge using > glm. Stata has no problem with convergence to the true parameters with the > same data. The simulated data are assumed to come from an underlying > latent variable. > > Thanks in advance for any insights, > Andrew > > ********* > > using DataFrames > using Distributions > using GLM > srand(10001) > > N=10000 > genX = Normal(0,3) > genɛ = Normal() > X = rand(genX,N) > ɛ = rand(genɛ,N) > α0 = 2 > α1 = -1 > Ystar = α0 + α1*X + ɛ > Y = (Ystar.>0)*1. > df = DataFrame(X=[X],Y=[Y]) > > Probit = glm(Y ~ X, df, Binomial(), ProbitLink()) >
