Modified Robust Method for Parameter Estimation of Multiple Regression Models
Article Sidebar
Main Article Content
Abstract
This research aimed to develop the parameter estimation by applying adjusted S-estimator and to compare the efficiency of parameter estimation by Mean Square Error (MSE) and the efficiency of forecasting by Mean Absolute Percentage Error (MAPE). The situations were simulated under 81 situations including 1) the error distributions are Standard Normal, Gamma and Weibull distribution, 2) the sample sizes are 20 60 and 100, 3) the percentage of outliers is 5, 25 and 40, and 4) the number of parameters is 2, 3 and 4. The research results demonstrated that: 1) the estimated parameters obtained from adjusted S-estimator are and unbiased parameters estimation of 2) The method of parameters estimation through adjusted S-estimator provided the value of MSE less than S-estimator from 59 situations. In most situations, the number of parameters is equal to 2 and 3 parameters. 3) The forecasting equation from the parameter estimation using adjusted S-estimator has the level of MAPE less than the level of MAPE of S-estimator from 41 situations which are mostly the Weibul distribution. The results of this research can be used as a guideline for selecting a robust parameter estimation of the multiple regression model when outliers occur with independent variables.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
The content and information in the article published in Journal of Rajamangala University of Technology Srivijaya It is the opinion and responsibility of the author of the article. The editorial journals do not need to agree. Or share any responsibility.
References
Ampanthong, P. and Suwattee, P. 2010. Robust estimation of regression coefficients with outliers. Thailand Statistician 8(2): 183-205. (in Thai)
Dave, M. B. and Nakrani, M. 2014. Malicious user detection in spectrum sensing for WRAN using different outliers detection techniques. Engineering Trends and Technology 9(7): 326-320.
Kamolsuk, N. 2020. Parameter Estimation for Robust Regression with Maximum Likelihood Estimation and S-estimation. Sripatum Review of Science and Technology 12(1): 190-203. (in Thai)
Montgomery, D., Peck, E. and Vining, G. 2006. Introductions to Linear Regression Analysis. 4th ed. John Wiley & Sons, NewYork.
Oller, V., Alfons, A. and Croux, C. 2016. The shooting S-estimator for robust regression. Computational Statistical 31(3): 829-844.
Panik, M. 2009. Regression Modeling Methods, Theory, and Computation with SAS. Taylor & Francis Group, New York.
Roelant, E., Aelst, S. and Croux, C. 2009. Multivariate generalized S-estimators. Multivariate Analysis 100(5): 876-887.
Rousseeuw, P.J. and Croux, C. 1993. Alternatives to the median absolute deviation. American Statistical Association 88(424): 1273-1283.
Rousseeuw, P.J. and Yohai, V. 1984. Robust Regression by Means of S-estimators, pp. 256-272. In Proceedings of Stochastische Mathematische Modelle. Heidelberg: University, Heidelberg.
Rousseeuw, P.J. and Leroy, A.M. 1987. Robust Regression and Outlier Detection. John wiley & sons, NewYork.
Rousseeuw, P.J. and Leroy, A. M. 2003. Robust Regression and Outlier Detection. John wiley & sons, New York.
Smirnov, P.O. and Shevlyakov, G.L. 2014. Fast highly efficient and robust one-step M-estimators of scale based on Qn. Computational Statistics and Data Analysis 78(4): 153-158.
Tabatabai, M., Ebay, H., Bae, S. and Singh, K. 2012. TELBS robust linear regression method. Open Access Medical Statistics 2(5): 65-84.
Tantrakul, O., Payakkapong, P. and Chomtee, B. 2012. Comparison of Robust Regression Methods in Multiple Linear Regression Model, pp. 291-300. In Proceeding of The 13th Graduate Research Conference. Khon Kan University, Khon Kan. (in Thai)
Tharmaratnam, K., Claeskens, G., Croux, C. and Salibián-Barrera, M. 2010. S-estimation for penalized regression splines. Computational and Graphical Statistics 19(3): 609-625.