ROBUST VARIABLE SELECTION FOR SINGLE INDEX SUPPORT VECTOR REGRESSION MODEL
DOI:
https://doi.org/10.23851/mjs.v30i1.388Keywords:
Single-index model, Support vector regression, Variable selection, High-dimensional, Outliers.Abstract
The single index support vector regression model (SI-SVR) is a useful regression technique used to alleviate the problem of high-dimensionality. In this paper, we propose a robust variable selection technique for the SI-SVR model by using vital method to identify and minimize the effects of outliers in the data set. The effectiveness of the proposed robust variable selection of the SI-SVR model is explored by using various simulation examples. Furthermore, the suggested method is tested by analyzing a real data set which highlights the utility of the proposed methodology.Metrics
References
Dhhan,W., Rana, S. and H. Midi (2015), Non-sparse ?-Insensitive Support Vector Regression for Outlier Detection
Journal of Applied Statistics, 42(8) (2015), 1723-1739;
Dhhan, W., Rana, S. Alshaybawee, T. and H. Midi "Elastic Net For Single Index Support Vector Regression Model." Economic Computation & Economic Cybernetics Studies & Research 51.2 (2017).
Guyon, I. & Elisseeff, A. (2003), An Introduction to Variable and Feature Selection. The Journal of Machine Learning Research, 3, 1157-1182;
Horowitz, J. L. (2009), Semiparametric and Nonparametric Methods in Econometrics. Springer;
Hu, Y., Gramacy, R. B. & Lian, H. (2013), Bayesian Quantile Regression for Single-index Models. Statistics and Computing, 23(4), 437-454;
Ichimura, H. (1993), Semiparametric Least Squares (SLS) and Weighted SLS Estimation of Single-Index Models. Journal of Econometrics, 58(1), 71-120;
Peng, H. & Huang, T. (2011), Penalized Least Squares for Single Index Models. Journal of Statistical Planning and Inference, 141(4), 1362-1379;
Wu, T. Z., Yu, K. & Yu, Y. (2010), Single-index Quantile Regression. Journal of Multivariate Analysis, 101(7), 1607-1621;
Downloads
Published
How to Cite
Issue
Section
License
Articles accepted for publication in Al-Mustansiriyah Journal of Science (MJS) are protected under the Creative Commons Attribution 4.0 International License (CC BY-NC). Authors of accepted articles are requested to sign a copyright release form prior to their article being published. All authors must agree to the submission, sign copyright release forms, and agree to be included in any correspondence between MJS and the authors before submitting a work to MJS. For personal or educational use, permission is given without charge to print or create digital copies of all or portions of a MJS article. However, copies must not be produced or distributed for monetary gain. It is necessary to respect the copyright of any parts of this work that are not owned by MJS.