Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | 유재근 | - |
| dc.contributor.author | 김서영 | - |
| dc.creator | 김서영 | - |
| dc.date.accessioned | 2025-02-18T16:32:23Z | - |
| dc.date.available | 2025-02-18T16:32:23Z | - |
| dc.date.issued | 2025 | - |
| dc.identifier.other | OAK-000000243264 | - |
| dc.identifier.uri | https://dcollection.ewha.ac.kr/common/orgView/000000243264 | en_US |
| dc.identifier.uri | https://dspace.ewha.ac.kr/handle/2015.oak/271005 | - |
| dc.description.abstract | This paper addresses the challenge of dimension reduction in datasets where both the predictor and response variables are multivariate. We introduce a novel approach that integrates the Minimum Average Variance Estimator (MAVE) with the Projective Resampling (PR) method to render its applicability in multivariate response settings. By applying MAVE within the framework of PR, we achieve an effective dimension reduction, ensuring the exhaustive recovery of the central mean subspaces and eradicating strong assumptions on the distribution of predictors. The proposed methodology is validated through a series of simulations to evaluate its effectiveness and robustness in various multivariate contexts. Further refinements were made to improve the computational performance as well. ;본 논문에서는 예측 변수와 반응 변수가 모두 다변량인 데이터셋에서 충분 차원 축소 문제를 다룹니다. 특히, Minimum Average Variance Estimator (MAVE)와 Projective Resampling (PR) 방법을 통합하여 다변량 반응 상황에 적용할 수 있는 새로운 접근법을 제안합니다. PR 프레임워크에서 MAVE를 활용함으로써 central mean subspace에 대해 exhaustive recovery를 수행하는 동시에, 예측 변수 분포에 대한 강한 가정을 없앰으로써 효율적인 차원 축소를 가능하게 합니다. 제안된 방법론의 효과와 견고성은 다양한 다변량 상황에서의 시뮬레이션 실험을 통해 검증되었습니다. 아울러 계산 성능을 향상시키기 위한 추가적인 개선 작업도 수행하였습니다. | - |
| dc.description.tableofcontents | I. Introduction 1 II. Literature Review on Classical Dimension Reduction Methods for Multivariate Response 5 A. Generalizing Slicing Scheme at Multidimensional Y. 5 B. Pooling the Marginal Dimension Reduction Space . 6 C. Narrowing the Target to the Central k-th Subspace . 6 III. Adaptive Estimation of Dimension Reduction . 8 A. Minimum Average Variance Estimator 8 B. Refined MAVE 11 IV. Projective Resampling and its Properties. 12 V. Projective Resampling in conjunction with the Minimum Average Variance Estimation 14 A. The Method 14 B. Algorithm 17 VI. Simulation. 19 VII. Discussion . 34 VIII. R-codes 35 Bibliography . 36 Abstract (in Korean) . 38 | - |
| dc.format | application/pdf | - |
| dc.format.extent | 1301075 bytes | - |
| dc.language | eng | - |
| dc.publisher | 이화여자대학교 대학원 | - |
| dc.subject | Dimension Reduction| Multivariate nonlinear regression| Central mean subspace| Forward regression | - |
| dc.subject.ddc | 500 | - |
| dc.title | Projective-resampling Minimum Average Variance Estimation for Multivariate Response | - |
| dc.type | Master's Thesis | - |
| dc.creator.othername | Kim, Seoyoung | - |
| dc.format.page | iv, 38 p. | - |
| dc.identifier.thesisdegree | Master | - |
| dc.identifier.major | 대학원 통계학과 | - |
| dc.date.awarded | 2025. 2 | - |