Feature selection in the semivarying coefficient LS-SVR

다국어 초록 (Multilingual Abstract)

In this paper we propose a feature selection method identifying important features in the semivarying coefficient model. One important issue in semivarying coefficient model is how to estimate the parametric and nonparametric components. Another issue is how to identify important features in the varying and the constant effects. We propose a feature selection method able to address this issue using generalized cross validation functions of the varying coefficient least squares support vector regression (LS-SVR) and the linear LS-SVR. Numerical studies indicate that the proposed method is quite effective in identifying important features in the varying and the constant effects in the semivarying coefficient model.

참고문헌 (Reference)

1 Hastie, T., "Varying-coeffcient models" 55 : 757-796, 1993

2 Shim, J., "Varying coeffcient modeling via least squares support vector regression" 161 : 254-259, 2015

3 Xue, L., "Variable selection in high-dimensional varying-coeffcient models with global optimality" 13 : 1973-1998, 2012

4 Vapnik, V., "The nature of statistical learning theory" Springer 1995

5 Tibshirani, R., "The lasso method for variable selection in the Cox model" 16 : 385-395, 1997

6 Fan, J., "Statistical methods with varying coeffcient models" 1 : 179-195, 2008

7 Vapnik, V., "Statistical Learning Theory" Wiley 1998

8 Craven, P., "Smoothing noisy data with spline functions: Estimating the correct degree of smoothing by the method of generalized cross-validation" 31 : 377-403, 1979

9 Li, Q., "Smooth varying-coeffcient estimation and inference for qualitative and quantitative data" 26 : 1607-1637, 2010

10 황창하, "Robust varying coe cient model using L1 regularization" 한국데이터정보과학회 27 (27): 1059-1066, 2016

11 Huang, J., "Regularized estimation in the accelerated failure time model with high dimensional covariates" Department of Statistics and Actuarial Science, The University of Iowa 2005

12 Fan, J., "Profile likelihood inferences on semiparametric varying coeffcient partially linear models" 11 : 1031-1057, 2005

13 Suykens, J. A. K., "Optimal control by least squares support vector machines" 14 : 23-35, 2001

14 Hoover, D. R., "Nonparametric smoothing estimates of time-varying coeffcient models with longitudinal data" 85 : 809-822, 1998

15 Zhang, W., "Local polynomial fitting in semivarying coeffcient models" 82 : 166-188, 2002

16 Suykens, J. A. K., "Least squares support vector machine classifiers" 9 : 293-300, 1999

17 Wooldridge, J. M., "Introductory econometrics: A modern approach" South-Western Cengage Learning 2012

18 Mercer, J., "Functions of positive and negative type and their connection with theory of integral equations" 415-446, 1909

19 Lee, Y. K., "Flexible generalized varying coeffcient regression models" 40 : 1906-1933, 2012

20 Yang, L., "Estimation and testing for varying coeffcients in additive models with marginal integration" 101 : 1212-1227, 2006

21 황창하, "Deep LS-SVM for regression" 한국데이터정보과학회 27 (27): 827-833, 2016

22 Wu, C., "A penalized robust semiparametric approach for gene-environment interactions" 34 : 4016-4030, 2015

23 Sauerbrei, W., "A bootstrap resampling procedure for model building : Application to the Cox regression model" 11 : 2093-2099, 1992