AbstractIn this paper, authors study properties and inference for the newly introduced skew-normal alpha-power model, generalizing both, the power-normal and skew-normal models. Inference is approached via maximum likelihood. Fisher information matrix is derived and shown to be nonsingular at the whole parametric space. Special emphasis is placed on the special case of the power–skew-normal model. Studies with real data illustrate the fact that the model can be very useful in applications, being able to overfit less general models entertained in the literature.Keywords: skew-normal distributionsmaximum likelihoodsingular information matrixView correction statement:Skew-normal alpha-power model [Statistics 48(2014) 1414–1428] AcknowledgementsThe authors acknowledge helpful comments and suggestions by the referee which substantially improved the presentation. The research of H. Bolfarine was supported by CNPq (Brazil). The research of H.W. Gómez was supported by FONDECYT (Chile) 1090411.