Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
5129818 | Statistics & Probability Letters | 2017 | 8 Pages |
Abstract
The feature screening procedure based on the expected conditional Kolmogorov filter is proposed for the ultrahigh dimensional binary classification problem with dependent variable. The sure screening and ranking consistency properties are established. Some numerical examples are also presented.
Related Topics
Physical Sciences and Engineering
Mathematics
Statistics and Probability
Authors
Peng Lai, Fengli Song, Kaiwen Chen, Zhi Liu,