DS+, BK 2023년도 제 15회 통계세미나 개최 안내(11/3(금))
DS플러스
2023-11-01
2023년도 제 15회 BK 통계 세미나 개최를 안내드립니다.
고려대학교 통계학과 통계연구소, BK21 통계학교육연구팀과 DS+ 사업단 주최로 이루어지는 세미나입니다.
일시 : 2023년 11월 3일 (금) 오후 5시
장소 : 고려대학교 정경관 206호
연사 : 김일문 교수 (연세대학교 응용통계학과)
주제 : Statistical Inference via Sample Splitting
Abstract : Classical asymptotic theory for statistical inference usually involves calibrating a statistic by fixing the dimension d while letting the sample size n increase to infinity. Recently, much effort has been dedicated towards understanding how these methods behave in high-dimensional settings, where d and n both increase to infinity simultaneously. This often leads to different inference procedures, depending on the assumptions about the dimensionality, leaving the practitioner in a bind. Motivated by this critical issue, we introduce a simple yet powerful approach whose validity does not depend on the assumption on d versus n. At the heart of our proposal are sample splitting and studentization that lead to a dimension-agnostic limiting distribution. In this talk, we exemplify our technique for a handful of classical and modern problems and discuss directions for future research.
홍보 자료 : 첨부파일 확인 부탁드립니다
[온라인 세미나 참여 링크]
ZOOM 링크
https://korea-ac-kr.zoom.us/j/81311119137?pwd=R0Z3SDM1ZG5kdzRzQnRkQ0cvSXNkdz09
- Zoom ID: 813 1111 9137
- Password: Kustat123@
앞으로는 해당 링크를 2학기동안 계속해서 사용할 예정입니다.
고려대학교 통계학과 통계연구소, BK21 통계학교육연구팀과 DS+ 사업단 주최로 이루어지는 세미나입니다.
일시 : 2023년 11월 3일 (금) 오후 5시
장소 : 고려대학교 정경관 206호
연사 : 김일문 교수 (연세대학교 응용통계학과)
주제 : Statistical Inference via Sample Splitting
Abstract : Classical asymptotic theory for statistical inference usually involves calibrating a statistic by fixing the dimension d while letting the sample size n increase to infinity. Recently, much effort has been dedicated towards understanding how these methods behave in high-dimensional settings, where d and n both increase to infinity simultaneously. This often leads to different inference procedures, depending on the assumptions about the dimensionality, leaving the practitioner in a bind. Motivated by this critical issue, we introduce a simple yet powerful approach whose validity does not depend on the assumption on d versus n. At the heart of our proposal are sample splitting and studentization that lead to a dimension-agnostic limiting distribution. In this talk, we exemplify our technique for a handful of classical and modern problems and discuss directions for future research.
홍보 자료 : 첨부파일 확인 부탁드립니다
[온라인 세미나 참여 링크]
ZOOM 링크
https://korea-ac-kr.zoom.us/j/81311119137?pwd=R0Z3SDM1ZG5kdzRzQnRkQ0cvSXNkdz09
- Zoom ID: 813 1111 9137
- Password: Kustat123@
앞으로는 해당 링크를 2학기동안 계속해서 사용할 예정입니다.