Alternative Expressions of Regression Vector for Principal Component Regression and Partial Least Squares Regression

• 발행기관: 한국자료분석학회
• 저자명: Jong-Duk Kim
• 간행물 정보: 『Journal of The Korean Data Analysis Society (JKDAS)』Vol.5 No.1, 17~26쪽, 전체 10쪽
• 주제분류: 자연과학 > 통계학
• 파일형태: PDF
• 발행일자: 2003.03.30

영문 초록

Principal component regression (PCR) and partial least squares regression (PLSR) are examined. They usually appear in completely different computational forms in the literature. Since PLSR procedure is always given as an iterative algorithm, it is difficult to see its relationship with PCR. In this paper, alternative unifying expressions of the regression vectors for PCR and PLSR are derived using singular value decomposition of the predictor variables matrix. They are non-iterative and comparative expressions and enable us to gain some insight into the interrelationships between those methods.

목차

1. Introduction

2. Development of Alternative Expressions

3. Comparative Expressions

4. Summary

References

키워드

Principal component regression Partial least squares regression Singular value decompositon Eigenvectors Krylov matrix