Stable feature screening for ultrahigh dimensional data

다국어 초록 (Multilingual Abstract)

This paper is concerned with the stable feature screening for the ultrahigh dimensional data. To deal with the ultrahigh dimensional data problem and screen the important features, a set-averaging measurement is proposed. The model averaging technique and the conditional quantile method are used to construct the weighted set-averaging feature screening procedure to identify the relationships between the possible predictors and the response variable. The proposed screening method is model free, stable and possesses the sure screening property under some regular conditions. Some Monte Carlo simulations and a real data application are conducted to evaluate the performance of the proposed procedure.

참고문헌 (Reference)

1 Hall, P., "Using generalized correlation to effect variable selection in very high dimensional problems" 18 : 533-550, 2009

2 Fan, J., "Sure independence screening for ultrahigh dimensional feature space" 70 : 849-911, 2008

3 Lin, B., "Stable prediction in high-dimensional linear models" 27 : 1401-1412, 2017

4 Li, G., "Robust rank correlation based screening" 40 : 1846-1877, 2012

5 Segal, M. R., "Regression approaches for microarray data analysis" 10 : 961-980, 2003

6 Ullah, A., "Parametric and nonparametric frequentist model selection and model averaging" 1 : 157-179, 2013

7 Liang, H., "Optimal weight choice for frequentist model average estimators" 106 : 1053-1066, 2011

8 Zhu, L., "Model-free feature screening for ultrahigh-dimensional data" 106 : 1464-1475, 2011

9 Lai, P., "Model free feature screening for ultrahigh dimensional data with responses missing at random" 105 : 201-216, 2017

10 Li, R., "Feature screening via distance correlation learning" 107 : 1129-1139, 2012

11 Zhang, J., "Correlation rank screening for ultrahigh-dimensional survival data" 108 : 121-132, 2017

12 Wu, Y., "Conditional quantile screening in ultrahigh-dimensional heterogeneous data" 102 : 65-76, 2015

13 Yuan, Z., "Combining linear regression models : When and how?" 100 : 1202-1214, 2005

14 Chen, X., "A note on quantile feature screening via distance correlation" 5 : 1-22, 2017

15 Ando, T., "A model-averaging approach for high-dimensional regression" 109 : 254-265, 2014