The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

Note on the estimation of informative predictor subspace and projective-resampling informative predictor subspace

다변량회귀에서 정보적 설명 변수 공간의 추정과 투영-재표본 정보적 설명 변수 공간 추정의 고찰

  • Yoo, Jae Keun (Department of Statistics, Ewha Womans University)
  • 유재근 (이화여자대학교 통계학과)
  • Received : 2022.08.17
  • Accepted : 2022.08.23
  • Published : 2022.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

An informative predictor subspace is useful to estimate the central subspace, when conditions required in usual suffcient dimension reduction methods fail. Recently, for multivariate regression, Ko and Yoo (2022) newly defined a projective-resampling informative predictor subspace, instead of the informative predictor subspace, by the adopting projective-resampling method (Li et al. 2008). The new space is contained in the informative predictor subspace but contains the central subspace. In this paper, a method directly to estimate the informative predictor subspace is proposed, and it is compapred with the method by Ko and Yoo (2022) through theoretical aspects and numerical studies. The numerical studies confirm that the Ko-Yoo method is better in the estimation of the central subspace than the proposed method and is more efficient in sense that the former has less variation in the estimation.

정보적 설명 변수 공간은 일반적인 충분차원축소 방법들이 요구하는 가정들이 만족하지 않을 때 중심부분공간을 추정하기 위해 유용하다. 최근 Ko와 Yoo (2022)는 다변량 회귀에서 Li 등 (2008)이 제시한 투영-재표본 방법론을 사용하여 정보적 설명 변수 공간이 아닌 투영-재표본 정보적 설명 변수 공간을 새로이 정의하였다. 이 공간은 기존의 정보적 설명 변수 공간에 포함되지만 중심 부분 공간을 포함한다. 본 논문에서는 다변량 회귀에서 정보적 설명 변수 공간을 직접적으로 추정할 수 있는 방법을 제안하고, 이를 Ko와 Yoo (2022)가 제시한 방법과 이론적으로 그리고 모의실험을 통해 비교하고자 한다. 모의실험에 따르면 Ko-Yoo 방법론이 본 논문에서 제시한 추정 방법보다 더 정확하게 중심 부분 공간을 추정하고, 추정값들의 변동이 적다는 측면에서 보다 더 효율적임을 알 수 있다.

Keywords

Acknowledgement

2019년도 정부(과학기술정보통신부)의 재원으로 한국연구재단의 지원을 받아수행된 기초연구사업임 (NRF2019R1F1A1050715).

References

  1. Ko S and Yoo JK (2022). Projective resampling estimation of informative predictor subspace for multivariate regression, Journal of the Korean Statistical Society, Available from: https://doi.org/10.1007/s42952-022-00178-0
  2. Lee K, Choi Y, Um HY, and Yoo JK (2019). On fused dimension reduction in multivariate regression, Chemometrics and Intelligent Laboratory Systems, 193, 103828. https://doi.org/10.1016/j.chemolab.2019.103828
  3. Li B, Wen S, and Zhu L (2008). On a projective resampling method for dimension reduction with multivariate responses, Journal of the American Statistical Association, 103, 1177-1186. https://doi.org/10.1198/016214508000000445
  4. Li KC (1991). Sliced inverse regression for dimension reduction, Journal of the American Statistcal Association, 86, 316-342. https://doi.org/10.1080/01621459.1991.10475035
  5. Park Y, Kim K, and Yoo JK (2022). On cross-distance selection algorithm for hybrid sufficient dimension reduction, Computational Statistics and Data Analysis, 176, 107562. https://doi.org/10.1016/j.csda.2022.107562
  6. Setodji CM and Cook RD (2004). K-means inverse regression, Technometrics, 46, 421-429. https://doi.org/10.1198/004017004000000437
  7. Yin X and Bura E (2006). Moment-based dimension reduction for multivariate response regression, Journal of Statististical Planning and Inference, 136, 3675-3688. https://doi.org/10.1016/j.jspi.2005.01.011
  8. Yoo JK and Cook RD (2007). Optimal sufficient dimension reduction for the conditional mean in multivariate regression, Biometrika, 943, 231-242.
  9. Yoo JK (2008). A novel moment-based sufficient dimension reduction approach in multivariate regresison, Com putational Statistics and Data Analysis, 52, 3843-3851. https://doi.org/10.1016/j.csda.2008.01.004
  10. Yoo JK (2016). Sufficient dimension reduction through informative predictor subspace, Statistics : A Journal of Theoretical and Applied Statistics, 50, 1086-1099.
  11. Yoo JK (2018). Tutorial: Dimension reduction in regression with a notion of sufficiency, Communications for Statistical Applications and Methods, 23, 93-103. https://doi.org/10.5351/CSAM.2016.23.2.093