论文部分内容阅读
This talk concerns with variable screening when highly correlated variables exist in high dimensional linear models.
Cluster Feature Selection in High Dimensional Linear Models
【摘 要】
：
This talk concerns with variable screening when highly correlated variables exist in high dimensional linear models.
【机 构】
：
The Hong Kong Polytechnic University,Hong Kong
【出 处】
：
上海交通大学
【发表日期】
：
2016年4期
其他文献
Recent advances in high-throughput biotechnologies have provided an unprecedented oppor-tunity for biomarker discovery,which,from a statistical point of view,can be cast as a variable selection proble
会议
会议
This talk will give an overview of the negative control approach for indirect adjustment of unobserved confounding in causal inference from observational data.
会议
To maximize the chance of finding outcome-associated variants,a number of consortia have been formed to conduct meta-analysis integrating results from multiple genome-wide association studies(GWAS).
Optimal Methods for DetectingWeak and Sparse Signals Based on Correlated Features in Genetic Associa
Optimal tests of detecting weak and sparse signals in big data,such as the Higher Criticism test(HC),Berk-Jones test(B-J),(o) -divergence tests have been shown extra statistical power in detecting nov
We consider the estimation of large covariance and precision matrices from high-dimensional sub-Gaussian observations with slowly decaying temporal dependence that is bounded by certain polynomial dec
会议
Increasing evidence has shown that epigenetic processes including methylation play fundamental roles in chronic diseases like cancer.
Local dependence is common in multivariate and object data sequences.We consider the testing and estimation of change-points in such sequences.
Colorectal Cancer Risk Classification with Genetic and Environmental Data via Kernel Machine Methods
Genetic studies of cancer diagnosis and progression have uncovered only a small number of risk markers explaining a small fraction of heritability and adding little improvement to disease risk predict