An Empirical Study on Dimension Reduction

• 발행기관: 한국자료분석학회
• 저자명: Changhee Suh Hakbae Lee
• 간행물 정보: 『Journal of The Korean Data Analysis Society (JKDAS)』Vol.20 No.6, 2733~2746쪽, 전체 14쪽
• 주제분류: 자연과학 > 통계학
• 파일형태: PDF
• 발행일자: 2018.12.31

영문 초록

The two inverse regression estimation methods, SIR and SAVE to estimate the central space are computationally easy and are widely used. However, SIR and SAVE may have poor performance in finite samples and need strong assumptions (linearity and/or constant covariance conditions) on predictors. The two non-parametric estimation methods, MAVE and dMAVE have much better performance for finite samples than SIR and SAVE. MAVE and dMAVE need no strong requirements on predictors or on the response variable. MAVE is focused on estimating the central mean subspace, but dMAVE is to estimate the central space. This paper explores and compares four methods to explain the dimension reduction. Each algorithm of these four methods is reviewed. Empirical study for simulated data shows that MAVE and dMAVE has relatively better performance than SIR and SAVE, regardless of not only different models but also different distributional assumptions of predictors. However, real data example with the binary response demonstrates that SAVE is better than other methods.

목차

1. Introduction

2. Dimension Reduction

3. Empirical Study

4. Conclusion

References

키워드

central space minimum average variance estimation (MAVE) density based on MAVE (dMAVE) sliced average variance estimation (SAVE) sliced inverse regression (SIR)