K. I. Santoro, H. J. Gómez, I. Barranco Chamorro, H. W. Gómez
Nonnegative data sets are quite usual in different scientific fields. To carry out inference for this kind of data, models of distributions with nonnegative support are required. In this sense, the Half-Normal, the Generalized Gamma, or the Truncated Positive Normal can be used. We also find models built on the basis of a given probability density function symmetric around zero, which exhibit good analytical properties. We follow this later approach, so the Extended Half-Power Exponential (EHPE) distribution which is based on the Power Exponential model. Its properties are discussed: cumulative distribution function, hazard function, moments, skewness and kurtosis coefficients. We highlight that the EHPE distribution can be expressed as a mixture of gamma distributions. Results are obtained from this fact. Estimation is carried out by maximum likelihood (ML) methods. A Monte Carlo simulation study shows the good performance of ML estimates. Real applications are included.
Keywords: Symmetric distributions, nonnegative distributions, kurtosis, maximum likelihood.
Scheduled
Statistical Models I
June 10, 2022 10:10 AM
A11
