The normality assumption is very attractive for the errors of regression models with continuous response variable. However, when it was not satisfied, some transformation can be adopted for the response variable to obtain, at least, the symmetry property. It is know that the estimates of the coefficients in normal regression models are sensitive to extreme observations. Alternatives to the assumption of normal errors have been proposed in the literature. One of those alternatives is to consider that the errors have distributions with heavier tails than the normal distribution, in order to reduce the influence of outlier observations. In this context, Lange et al. (J Am Stat Assoc 84:881–896, 1989) proposed the Student t model with unknown ν degrees of freedom. In the last decade, several results appeared as alternatives to modeling other distributions than the normal errors as, for instance, the symmetrical (or elliptical) distributions. Some of these results can be found in Fang et al., Symmetric Multivariate and Related Distributions. Chapman and Hall, London (1990) and Fang and Anderson, Statistical Inference in Elliptical Contoured and Related Distributions. Allerton Press, New York (1990).