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.