In [Martínez-Flórez G, Bolfarine H, Gómez HW. Skew-normal alpha-power model. Statistics. 2014;48(6):1414–1428] the authors make an error in calculating the information matrix of the skew-normal alpha-power model under the normality hypothesis H0:(λ,α)=(0,1). Here we prove that the information matrix under H0 is singular and we proposed a reparametrization that allows one to calculate the asymptotic distribution of the maximum likelihood estimator.
Tópico:
Statistical Distribution Estimation and Applications
