Published

2022-01-01

Likelihood-Based Inference for the Asymmetric Exponentiated Bimodal Normal Model

Inferencia basada en verosimilitud para el modelo asimétrico bimodal normal exponenciado

DOI:

https://doi.org/10.15446/rce.v45n2.95530

Keywords:

Alpha-Power distribution, Asymmetric models, Bimodal normal distribution, Censored data, Maximum likelihood estimation (en)
Distribución normal bimodal, Distribución alfa-potencia, Datos censurados, Modelos asimétricos, Estimación por máxima verosimilitud (es)

Downloads

Authors

  • Guillermo Martínez-Flórez Departamentode Matemáticas y Estadística, Facultadde Ciencias Básicas, Universidad de Córdoba, Montería, Colombia
  • Mario Pacheco-López Facultad de Estadística, Universidad Santo Tomás, Bogotá, Colombia
  • Roger Tovar-Falón Departamento de Matemáticas y Estadística, Universidad de Córdoba, Montería, Colombia

Asymmetric probability distributions have been widely studied by various authors in recent decades, who have introduced new families of flexible distributions in terms of skewness and kurtosis than the classical distributions known in statistical theory. Most of the new distributions fit unimodal data, others fit bimodal data, however, in the bimodal, singularity problems have been found in their information matrices in most of the proposals presented. In contrast, in this paper an extension of the family of alpha-power distributions was developed, which has a non-singular information matrix, based on the bimodal-normal and bimodal elliptic-skew-normal probability distributions. These new extensions model asymmetric bimodal data commonly found in various areas of scientific interest. The properties of these new probabilistic distributions were also studied in detail and the respective statistical inference process was carried out to estimate the parameters of these new models. The stochastic convergence for the vector of maximum likelihood estimators could be found due to the non-singularity of the expected information matrix in the corresponding support.

Las distribuciones de probabilidad asimétricas han sido ampliamente estudiadas por varios autores en las últimas décadas, quienes han introducido nuevas familias de distribuciones flexibles en términos de asimetría y curtosis que las distribuciones clásicas conocidas en teoría estadística. La mayoría de las nuevas distribuciones se ajustan a datos unimodales, otras se ajustan a datos bimodales, sin embargo, en el caso bimodal, se han encontrado problemas de singularidad en la matriz de información en la mayoría de las propuestas presentadas. En contraste, en este trabajo se desarrolló una extensión de la familia de distribuciones de alfa-potencia, la cual tiene una matriz de información no singular, basada en las distribuciones de probabilidad normal-bimodal y bimodal elíptica-sesgada-normal. Estas nuevas extensiones modelan datos bimodales asimétricos que se encuentran comúnmente en diversas áreas de interés científico. También se estudiaron en detalle las propiedades de estas nuevas distribuciones probabilísticas y se llevó a cabo el respectivo proceso de inferencia estadística para estimar los parámetros de estos nuevos modelos. La convergencia estocástica para el vector de estimadores de máxima verosimilitud se pudo encontrar debido a la no singularidad de la matriz de información esperada en el soporte correspondiente.

References

Akaike, H. (1974), 'A new look at the statistical model identification', IEEE Transactions on Automatic Control 19(6), 716-723.

Arnold, B. C., Gómez, H. & Salinas, H. (2009), 'On multiple contraint skewed models', Scandinavian Journal of Statistics 43(3), 279-293.

Azzalini, A. (1985), 'A class of distributions which includes the normal ones', Scandinavian Journal of Statistics 12(2), 171-178.

Azzalini, A. (1986), 'Further results on a class of distributions which includes the normal ones', Statistical 46(2), 199-208.

Azzalini, A. & Bowman, A. (1990), 'A look at some data on the old faithful geyser', Journal of the Royal Statistical Society. Series C (Applied Statistics) 39(3), 357-365.

Bolfarine, H., Martínez-Flórez, G. & Salinas, H. (2018), 'Bimodal symmetric-asymmetric power-normal families', Communications in Statistics-Theory and Methods 47(2), 259-276.

Bozdogan, H. (1987), 'Model selection and Akaike's information criterion (AIC): The general theory and its analytical extensions', Psychometrika 52(3), 345-370.

Durrans, S. (1992), 'Distributions of fractional order statistics in hydrology', Water Resources Research 28(6), 1649-1655.

Elal-Olivero, D. (2010), 'Alpha-skew-normal distribution', Proyecciones Journal of Mathematics 29(3), 224-240.

Elal-Olivero, D., Gómez, H. W. & Quintana, F. A. (2009), 'Bayesian modeling using a class of bimodal skew-elliptical distributions', Journal of Statistical Planning and Inference 139(4), 1484-1492.

Elal-Olivero, D., Olivares-Pacheco, J., Gómez, H. & Bolfarine, H. (2009), 'A new class of non negative distributions generated by symmetric distributions', Communications in Statistics - Theory and Methods 38(7), 993-1008.

Gómez, H., Olivero, D., Salinas, H. & Bolfarine, H. (2009), 'Bimodal extensión based on the skew-normal distribution with application to pollen data', Environmetrics 22(1), 50-62.

Gupta, R. & Gupta, R. (2004), 'Generalized skew normal model', Test 13, 501-524.

Gupta, R. & Gupta, R. (2008), 'Analyzing skewed data by power normal model', Test 17, 197-210.

Hartigan, J. & Hartigan, P. (1985), 'The dip test of unimodality', The Annals of Statistics 13(1), 70-84.

Hartigan, P. (1985), 'Algorithm AS 217: Computation of the dip statistics to test for unimodality', Journal of the Royal Statistical Society. Series C (Applied Statistics) 34(1), 320-325.

Henze, N. (1986), 'A probabilistic representation of the skew-normal distribution', Scandinavian Journal of Statistics 13(4), 271-275.

Kim, H. (2005), 'On a class of two-piece skew-normal distribution', Statistics 39(6), 537-553.

Marin, J., Mengersen, K. & Robert, C. (2005), 'Bayesian modelling and inference on mixtures of distributions', Handbook of Statistics 25, 459-507.

Martínez-Flórez, G., Bolfarine, H. & Gómez, H. (2014), 'Skew-normal alpha-power model', Statistics 48(6), 1414-1428.

Martínez-Flórez, G., Bolfarine, H. & Gómez, H. (2018), 'Censored bimodal symmetric-asymmetric families', Statistics and Its Interface 11(2), 237-249.

Martínez-Flórez, G., Tovar-Falón, R. & Jiménez-Narváez, M. (2020), 'Likelihood-Based Inference for the Asymmetric Beta-Skew Alpha-Power Distribution', Symmetry 12(4), 613.

Pewsey, A. (2000), 'Problems of inference for Azzalini's skew-normal distribution', Journal of Applied Statistics 27(7), 859-870.

Pewsey, A., Gómez, H. & Bolfarine, H. (2012), 'Likelihood-based inference for power distributions', Test 21, 775-789.

R Development Core Team (2021), R: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0. http://www.R-project.org

Schwarz, G. E. (1978), 'Estimating the dimension of a model', Annals of Statistics6(2), 461-464.

Sulewski, P. (2019), 'Modified Lilliefors goodness-of-fit test for normality', Communications in Statistics-Simulation and Computation. doi.org/10.1080/03610918.2019.1664580

Torabi, H., Montazeri, N. H. & Grané, A. (2016), 'A test for normality based on the empirical distribution function', SORT 40(1), 55-88.

Dimensions

PlumX

Article abstract page views

673

Downloads

Download data is not yet available.

How to Cite

Martínez-Flórez, G., Pacheco-López, M., & Tovar-Falón, R. (2022). Likelihood-Based Inference for the Asymmetric Exponentiated Bimodal Normal Model. Revista Colombiana De Estadística, 45(2), 301-326. https://doi.org/10.15446/rce.v45n2.95530