Published
Bimodal Regression Model
Modelo de regresión Bimodal
DOI:
https://doi.org/10.15446/rce.v40n1.51738Keywords:
Bimodal distribution, Generalized Gaussian distribution, Linear regression, Power normal model, Regression (en)distribución bimodal, distribución gaussiana generalizada, regresión lineal, modelo de regresión exponenciado. (es)
El análisis de regresión es una técnica muy utilizada en diferentes áreas de conocimiento humano, con diferentes distribuciones para el término de error, sin embargo los modelos de regresión con el termino de error siguiendo una distribución bimodal no son comunes en la literatura, tal vez por la simple razón de no tratar con errores con distribución bimodal. En este trabajo proponemos un camino sencillo para hacer frente a modelos de regresión bimodal con una distribución simétrica - asimétrica para el término de error para la cual para algunos valores del parámetro de forma esta puede ser bimodal. Esta nueva distribución contiene a la distribución normal y la distribución normal asimétrica como casos especiales. Una aplicación con datos reales muestra que el nuevo modelo puede ser extremadamente útil en algunas situaciones.
https://doi.org/10.15446/rce.v40n1.51738
1Universidad de Córdoba, Facultad de Ciencias Básicas, Departamento de Matemáticas y Estadística, Córdoba, Colombia. PhD. Email: gmartinez@correo.unicordoba.edu.co
2Universidad de Atacama, Facultad de Ingenieróia, Departamento de Matemóatica, Copiapó, Chile. PhD. Email: hugo.salinas@uda.cl
3Universidad de Sao Paulo, IME, Departamento de Estatóistica, Sao Paulo, Brasil. PhD. Email: hbolfar@ime.usp.br
Regression analysis is a technique widely used in different areas of human knowledge, with distinct distributions for the error term. It is the case, however, that regression models with the error term following a bimodal distribution are not common in the literature, perhaps due to the lack of simple to deal with bimodal error distributions. In this paper, we propose a simple to deal with bimodal regression model with a symmetric-asymmetric distribution for the error term for which for some values of the shape parameter it can be bimodal. This new distribution contains the normal and skew-normal as special cases. A real data application reveals that the new model can be extremely useful in such situations.
Key words: Bimodal Distribution, Generalized Gaussian Distribution, Linear Regression, Power Regression Model.
El análisis de regresión es una técnica muy utilizada en diferentes áreas de conocimiento humano, con diferentes distribuciones para el término de error, sin embargo los modelos de regresión con el termino de error siguiendo una distribución bimodal no son comunes en la literatura, tal vez por la simple razón de no tratar con errores con distribución bimodal. En este trabajo proponemos un camino sencillo para hacer frente a modelos de regresión bimodal con una distribución simétrica - asimétrica para el término de error para la cual para algunos valores del parámetro de forma esta puede ser bimodal. Esta nueva distribución contiene a la distribución normal y la distribución normal asimétrica como casos especiales. Una aplicación con datos reales muestra que el nuevo modelo puede ser extremadamente útil en algunas situaciones.
Palabras clave: distribución bimodal, distribución gaussiana generalizada, regresión lineal, modelo de regresión exponenciado.
Texto completo disponible en PDF
References
1. Arellano-Valle, R. B., Bolfarine, H. & Vilca-Labra, F. (1996), 'Ultrastructural elliptical models', The Canadian Journal of Statistics 24(2), 207-216.
2. Azzalini, A. (1985), 'A class of distributions which includes the normal ones', Scandinavian Journal of Statistics 12(2), 171-178.
3. Bolfarine, H., Martínez-Flórez, G. & Salinas, H. S. (2012), 'Bimodal symmetric-asymmetric power-normal families', Communications in Statistics-Theory and Methods. DOI:10.1080/03610926.2013.765475.
4. Cancho, V. G., Lachos, V. H. & Ortega, E. M. M. (2010), 'A nonlinear regression model with skew-normal errors', Statistical Papers 51(3), 547-558.
5. Casella, G. & Berger, R. (2002), Statistical Inference, 2 edn, Thomson, Singapore.
6. Cordeiro, G. M., Ferrari, S. L. P., Uribe-Opazo, M. A. & Vasconcellos, K. L. P. (2000), 'Corrected maximum likelihood estimation in a class of symmetric nonlinear regression models', Statistics & Probability Letters 46(4), 317-328.
7. De Veaux, R. (1989), 'Mixtures of linear regressions', Computational Statistics and Data Analysis 8, 227-245.
8. Durrans, S. R. (1992), 'Distributions of fractional order statistics in hydrology', Water Resources Research 28(6), 1649-1655.
9. Galea, M., Paula, G. A. & Cysneiros, J. A. (2005), 'On diagnostic in symmetrical nonlinear models', Statistics & Probability Letters 73(4), 459-467.
10. Gupta, R. D. & Gupta, R. C. (2008), 'Analyzing skewed data by power normal model', TEST 17(1), 197-210.
11. Lange, K. L., Little, R. J. A. & Taylor, J. M. G. (1989), 'Robust Statistical Modeling Using the t Distribution', Journal of the American Statistical Association 84(408), 881-896.
12. Lehman, E. L. & Casella, G. (1998), Theory of Point Estimation, 2 edn, Springer, New York.
13. Lin, G. & Stoyanov, J. (2009), 'The logarithmic Skew-Normal Distributions are Moment-Indeterminate', Journal of Applied Probability Trust 46, 909-916.
14. Martínez-Flórez, G., Bolfarine, H. & Gómez, H. W. (2015), 'Likelihood-based inference for the power regression model', SORT 39(2), 187-208.
15. Pewsey, A., Gómez, H. W. & Bolfarine, H. (2012), 'Likelihood based inference for distributions of fractional order statistics', TEST 21(4), 775-789.
16. Quandt, R. (1958), 'The estimation of the parameters of a linear regression system obeying two separate regimes', Journal of the American Statistical Association 53(284), 873-880.
17. R Core Team, (2015), 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/}
18. Rego, L. C., Cintra, R. J. & Cordeiro, G. M. (2012), 'On some properties of the beta normal distribution', Communications in Statistics-Theory and Methods 41, 3722-3738.
19. Therneau, T., Grambsch, P. & Fleming, T. (1990), 'Martingale-based residuals for survival models', Biometrika 77, 147-160.
20. Turner, T. (2000), 'Estimating the propagation rate of a viral infection of potato plants via mixtures of regressions', Applied Statistics 49(3), 371-384.
21. Young, D. S. & Hunter, D. R. (2010), 'Mixtures of regressions with predictor-dependent mixing proportions', Computational Statistics and Data Analysis 54(10), 2253-2266.
Este artículo se puede citar en LaTeX utilizando la siguiente referencia bibliográfica de BibTeX:
@ARTICLE{RCEv40n1a03,
AUTHOR = {Martínez-Flórez, Guillermo and Salinas, Hugo S. and Bolfarine, Heleno},
TITLE = {{Bimodal Regression Model}},
JOURNAL = {Revista Colombiana de Estadística},
YEAR = {2017},
volume = {40},
number = {1},
pages = {65-83}
}
References
Arellano-Valle, R. B., Bolfarine, H. & Vilca-Labra, F. (1996), ‘Ultrastructural elliptical models’, The Canadian Journal of Statistics 24(2), 207–216.
Azzalini, A. (1985), ‘A class of distributions which includes the normal ones’, Scandinavian Journal of Statistics 12(2), 171–178.
Bolfarine, H., Martínez-Flórez, G. & Salinas, H. S. (2012), ‘Bimodal symmetricasymmetric power-normal families’, Communications in Statistics-Theory and Methods . DOI:10.1080/03610926.2013.765475.
Cancho, V. G., Lachos, V. H. & Ortega, E. M. M. (2010), ‘A nonlinear regression model with skew-normal errors’, Statistical Papers 51(3), 547–558.
Casella, G. & Berger, R. (2002), Statistical Inference, 2 edn, Thomson, Singapore.
Cordeiro, G. M., Ferrari, S. L. P., Uribe-Opazo, M. A. & Vasconcellos, K. L. P. (2000), ‘Corrected maximum likelihood estimation in a class of symmetric nonlinear regression models’, Statistics & Probability Letters 46(4), 317–328.
De Veaux, R. (1989), ‘Mixtures of linear regressions’, Computational Statistics and Data Analysis 8, 227–245.
Durrans, S. R. (1992), ‘Distributions of fractional order statistics in hydrology’, Water Resources Research 28(6), 1649–1655.
Galea, M., Paula, G. A. & Cysneiros, J. A. (2005), ‘On diagnostic in symmetrical nonlinear models’, Statistics & Probability Letters 73(4), 459–467.
Gupta, R. D. & Gupta, R. C. (2008), ‘Analyzing skewed data by power normal model’, TEST 17(1), 197–210.
Lange, K. L., Little, R. J. A. & Taylor, J. M. G. (1989), ‘Robust Statistical Modeling Using the t Distribution’, Journal of the American Statistical Association 84(408), 881–896.
Lehman, E. L. & Casella, G. (1998), Theory of Point Estimation, 2 edn, Springer, New York.
Lin, G. & Stoyanov, J. (2009), ‘The logarithmic Skew-Normal Distributions are Moment-Indeterminate’, Journal of Applied Probability Trust 46, 909–916.
Martínez-Flórez, G., Bolfarine, H. & Gómez, H. W. (2015), ‘Likelihood-based inference for the power regression model’, SORT 39(2), 187–208.
Pewsey, A., Gómez, H. W. & Bolfarine, H. (2012), ‘Likelihood based inference for distributions of fractional order statistics’, TEST 21(4), 775–789.
Quandt, R. (1958), ‘The estimation of the parameters of a linear regression system obeying two separate regimes’, Journal of the American Statistical Association 53(284), 873–880.
R Core Team (2015), R: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-
Rego, L. C., Cintra, R. J. & Cordeiro, G. M. (2012), ‘On some properties of the beta normal distribution’, Communications in Statistics-Theory and Methods 41, 3722–3738.
Therneau, T., Grambsch, P. & Fleming, T. (1990), ‘Martingale-based residuals for survival models’, Biometrika 77, 147–160.
Turner, T. (2000), ‘Estimating the propagation rate of a viral infection of potato plants via mixtures of regressions’, Applied Statistics 49(3), 371–384.
Young, D. S. & Hunter, D. R. (2010), ‘Mixtures of regressions with predictor-dependent mixing proportions’, Computational Statistics and Data Analysis 54(10), 2253–2266.
How to Cite
APA
ACM
ACS
ABNT
Chicago
Harvard
IEEE
MLA
Turabian
Vancouver
Download Citation
CrossRef Cited-by
1. Ahmad Abu-Akel, Carrie Allison, Simon Baron-Cohen, Dietmar Heinke. (2019). The distribution of autistic traits across the autism spectrum: evidence for discontinuous dimensional subpopulations underlying the autism continuum. Molecular Autism, 10(1) https://doi.org/10.1186/s13229-019-0275-3.
2. Mijeong Kim. (2023). Appropriate use of parametric and nonparametric methods in estimating regression models with various shapes of errors. Stat, 12(1) https://doi.org/10.1002/sta4.606.
Dimensions
PlumX
Article abstract page views
Downloads
License
Copyright (c) 2017 Revista Colombiana de Estadística
This work is licensed under a Creative Commons Attribution 4.0 International License.
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).