[R] convergence warning in betamix()

Hi, I am running some rather complex mixtures of beta regressions using the betamix() command from the betareg package (V. 3.0-4). If I am doing exploratory regressions with only one random starting value (nstart=1) I obtain results which converge after about 100 iterations. However, if I run regre

Re: [R] ICL for betamix()

I found a way to manually calculate the ICL for a betamix() result, using the entropy.empirical() command from the 'entropy' package. Here is the code: model=betamix(y~x) BIC(model)+entropy.empirical(posterior(model)) Cheers, Chris -- View this message in context: http://r.789695.n4.nabble.co

[R] ICL for betamix()

Hi, I am running beta mixture regressions using the betamix() command from the package 'betareg' (3.0-4). I In order to inform the choice about the number of latent classes I took a look at the various information criteria (AIC, BIC, ICL) and learnt that the integrated completed likelihood (ICL) i

[R] logLik in betareg()

Hi everyone, I have estimated different models with the betareg() command from the package 'betareg' (3.0-4). When I started to compare them using likelihood ratio tests, it occured to me that the logLik() of the models increased with increasing number of parameters. I confirmed this observations

[R] hetglm() and robust standard errors

Hi everyone, I am using the hetglm() command from the package 'glmx' (0.1-0). It seems that hetglm() is incompatible with the robust standard errors estimator provided in the 'AER' package: coeftest(mymodel,vcov=vcovHC) Any suggestions how I could obtain robust standard errors for the heteroscedast