Published
2015-12-30

Comparación de la regresión GINI con la regresión de mínimos cuadrados ordinarios y otros modelos de regresión lineal robustos

Comparison of Gini Regression with OLS Regression and other Robust Linear Regression

DOI: https://doi.org/10.15332/s2027-3355.2015.0002.01
Juan Carlos Correa Morales
Universidad Nacional de Colombia Facultad de Ciencias, Escuela de Estadística. Colombia
Gloria Patricia Carmona Florez
Universidad Nacional de Colombia Sede Medellín
PDF (Spanish) BibTex (Spanish)

Abstract (en)

In this paper compares Gini regression (using the non-parametric approach of weighted average of slope, instead of the parametric approach) with OLS and other robust regression methods, the type L (LAV, linear combinations of order statistics), the type M (M Huber, based on the concept of maximum likelihood) and the type MM (based on the minimization of an estimator M). The comparison of the methods is performed via simulation under different scenarios. The results show that the Gini regression has a higher degree of robustness compared with the OLS regression to estimate the regression coefficients in the presence of outliers, but their robustness is less than robust estimation methods LAV, M Huber and MM.
Keywords (en): Datos atípicos, eficiencia, míminos cuadrados ordinarios, modelos de regresión robustos, regresión Gini, robustez.

Abstract (es)

En este trabajo se compara la regresión de Gini con la regresión OLS y otros métodos de regresión robustos, del tipo L (LAV, combinaciones lineales de estadísticos de orden), del tipo M (M de Huber, basado en el concepto de máxima verosimilitud) y del tipo MM (basado en la minimización de un estimador M). La comparación de los métodos se realiza vía simulación bajo diferentes escenarios. Los resultados encontrados vía simulación muestran que la regresión de Gini tiene un mayor grado de robustez en comparación con la regresíon OLS  al estimar los coeficientes de regresión ante la presencia de datos atípicos, pero su robustez es menor que la de los métodos de estimación robustos LAV, M de Huber y MM
Keywords (es): Datos atípicos, eficiencia, míminos cuadrados ordinarios, modelos de regresión robustos, regresión Gini, robustez.
Gloria Patricia Carmona Florez, Universidad Nacional de Colombia Sede Medellín

Antioquia

Analista

References

Aelst, V. S., W. G. & Zamar, H. R. (2013), ‘Robust and efficient estimation of the residual scale in linear regression.’, Journal of Multivariate Analysis 116, 278–296.

Bassett, G., J. & Koenker, R. (1978), ‘Asymptotic theory of least absolute error regression.’, Journal of the American Statistical Association 73, 618–622.

Bianco, M. A., G. M. B. y. Y. J. (2005), ‘Robust estimation for linear regression with asymmetric errors.’, The Canadian Journal of Statistics 33, 511–528.

Birkes, D. & Dodge, Y. (1993), Alternative Methods of Regression, 1 edn, John Wiley & Sons Inc., New York.

Boente, G. & Fraiman, R. (1989), ‘Robust Nonparametric Regression Estimation.’, Journal of Multivariate Analysis 29, 180–198.

Borroni, C. G. & Cazzaro, M. (2006), ‘Some developments about a new nonparametric test based on Gini’s mean difference.’, Statistica & Applicazioni 3, 29–44.

Borroni, C. & Zenga, M. (2006), ‘A test of concordance based on Gini’s mean difference.’, Communications in Statistics - Theory and Methods 16, 289–308.

Boscovich, R. (1757), ‘De literaria expeditione per pontificiam ditioned. et. synopsis amplioris opens, ac habeniur plura ejus ex exemplaria etiam sensorium impressa.’, Bononiensi Scientiarum el Anum Instituto Atque Academia Commentarii 4, 353–396.

Casella, G. & Berger, R. L. (2002), Statistical Inference, 2 edn, Duxbury, New York.

David, H. A. (1968), ‘Gini’s Mean Difference Rediscovered.’, Biometrika 55, 573–575.

Denby, L., a. L. W. A. (1977), ‘Robust Regression Estimators Compared via Monte Carlo.’, Communications in Statistics-Theory and Methods 6, 335–362.

Douglas, C., M. E. A. P. & Vining, G. G. (2002), Introducción al análisis de regresión lineal, 2 edn, John Wiley & Sons, New York.

E. Schechtman, E., Y. S. (2003), ‘A Family of Correlation Coefficients Based on Extended Gini.’, Journal of Economic Inequality 1, 3083–3088.

Edna, S., Y. S. & Artsev, Y. (2008), ‘ Who Does Not Respond in the Household Expenditure Survey: An Exercise in Extended Gini Regressions.’, American

Statistical Association Journal of Business & Economic Statistics 26, 329– 344.

Edna, S., Y. S. & Pudalov, T. (2011), ‘Gini’s multiple regressions: two approaches and their Interaction.’, International Journal of Statistics 69, 67–99.

Edna, S. & Yitzhaki, S. (2007), ‘ A Measure Of Association Based On Gini’s Mean Difference.’, Communications in Statistics-Theory and Methods 16, 207–231.

Edna, S. & Yitzhaki, S. (2013), Gini’s Multiple Regressions, 1 edn, John Wiley & Sons, New York.

Hampel, F. R., R. E. M. R. P. J. & Stahel, W. A. (1986), Robust statistics: The approach based on inuence functions., 1 edn, Wiley, New York.

Hettmansperger, T. P. (1984), Statistical inference based on ranks., 1 edn, John Wiley & Sons, New York.

Huber, P, J. (1973), ‘ Regression: Asymptotics, Conjectures, and Monte Carlo.’, Annals of Statistics 1, 799–821.

Huber, P. J. (1981), Robust statistics., 1 edn, John Wiley & Sons, New York.

Hurdle, W. (1984), ‘Robust Regression Function Estimation.’, Journal of Multivariate Analysis 14, 169–180.

Iachan, R. (1985), ‘Robust Designs for Ratio and Regression Estimation.’, Journal of Statistical Planning and Inference 11, 149–161.

T.E., D. (2005), ‘Least Absolute Value Regression: Recent Contributions.’, Journal of Statistical Computation and Simulation 75, 263–286.

Dimensions

PlumX

Visitas

1658

Downloads

How to Cite

Correa Morales, J. C., & Florez, G. P. C. (2015). Comparison of Gini Regression with OLS Regression and other Robust Linear Regression. Comunicaciones En Estadística, 8(2), 129-161. https://doi.org/10.15332/s2027-3355.2015.0002.01

License

The authors maintain the rights to the articles and therefore they are free to share, copy, distribute, execute and publicly communicate the work under the following conditions:

Recognize the credits of the work in the manner specified by the author or licensor (but not in a way that suggests that, you have their support or that they support your use of their work).

Comunicaciones en Estadística is licensed under Creative Commons Atribución-NoComercial-CompartirIgual 4.0 Internacional (CC BY-NC-SA 4.0)

Universidad Santo Tomás preserves the patrimonial rights (copyright) of the published works, and favors and allows the reuse of them under the aforementioned license.