A new package gamlss.inf is introduced recently to help with the fitting of inflated distributions on the interval [0,1] and zero adjusted distributions on the interval (o, infinity) for a response variable.

The inflated and zero adjusted distributions are mixed (i.e continuous and discrete) distributions with a continuous part on a real line interval and an additional part with point probabilities at either 0 or 1 or both 0 and 1.

The gamlss package gamlss.dist already provides the inflated beta distributions, BEINF, BEINF0, BEINF1 BEZI and BEOI which allow the user to fit a beta distribution on (0,1), with extra point probabilities at 0 or 1 or both 0 and 1. The probabilities at the points 0 and 1 may depend on explanatory variables. Since the beta distribution has 2 parameters, the inflated beta, BEINF, (with the addition of the two probabilities at 0 and 1) has a total of 4 parameters. In practice, and for complicated data sets, the part of the response which lies on (0,1) may need more than 2 distribution parameters to be captured correctly. The vignette Inflated distributions on the interval [0, 1] provides information how this can be done.

The gamlss package gamlss.dist also has two zero adjusted distributions ZAGA and ZAIG (gamma and inverse Gaussian; respectively). The package gamlss.inf allows all possible gamlss.family distributions defined on (o, infinity) to be zero adjusted. For more information see the vignette Zero adjusted distributions on the positive real line.

Hi Gamlss team,

I have a problem using random effect in function gamlssZadj(). Could you please help me about the error message? Thank you!

#gamlss() works

> test test test

Did you use random() or re()

random() should work

Mikis