NEYMAN-PEARSON CLASSIFICATION UNDER HIGH-DIMENSIONAL SETTINGS

来源 :泛华统计学会(icsa)2015年学术会议 | 被引量 : 0次 | 上传用户：CrazyDesire

【摘要】

：

Most existing binary classification methods target on the opti-mization of the overall classification risk and may fail to serve some real-world applications su

【作者】

：

Xin Tong

【机构】

：

UniversityofSouthernCalifornia,USA

【出处】

：

泛华统计学会(icsa)2015年学术会议

【发表日期】

：

2015年7期

下载到本地 , 更方便阅读

下载此文赞助VIP

声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架

论文部分内容阅读

Most existing binary classification methods target on the opti-mization of the overall classification risk and may fail to serve some real-world applications such as cancer diagnosis,where users are more concerned with the risk of misclassifying one specific class than the other.

其他文献

ROC for incident disease using multivariate markers with nonrandom dropout

Although variance is usually dealt as a nuisance in many classical statistical analyses,it has received increasing attention as an interest of its own in many f

会议

Intrinsic random functions and universal kriging on the circle

Intrinsic random functions(IRF) provide a versatile approach when the assumption of second-order stationarity is not met.Here,we develop the IRF theory on the c

会议

A semiparametrically efficient estimator of the time-varying effects for survival data with time-dep

会议

Statistical Methods for Studying Disease Etiologic Heterogeneity

A fundamental goal of epidemiologic research is to investigate the relationship between exposures and disease risk.Cases of the disease are often considered a s

会议

Power Calculations for the Stepped Wedge Designs

Background:The stepped wedge design(SWD) provides a rigorous,randomized scheme for evaluation of interventions.One example is the assessment of early ART access

会议

Sequential Probability Ratio Tests in Medical Product Safety Evaluation

Sequential probability ratio test(SPRT) and its variants have recently received much attention in safety evaluation due to its some well-known merits,e.g.,minim

会议

A Pairwise-Likelihood Augmented Estimator for Cox Model Under Left-Truncation

Survival data collected from prevalent cohorts are subject to left-truncation.Conventional conditional approaches can be inefficient by ignoring the information

会议

Coauthorship and Citation networks for Statisticians

We have collected and cleaned two network data sets:Coauthorship and Citation networks for statisticians.The data sets are based on all research papers publishe

会议

Latent Variable Joint Model for Multivariate Longitudinal and Survival Data

　　This study proposes a latent variable framework for jointly analyzing multivariate longitudinal data with time to event survival outcome.Latent variable mix

会议

latent variable modelsurvival analysismultivariate longitudinal data

An Adaptive Combination Method for Averaging OLS and GLS Estimators

To avoid the risk of misspecification between homoscedastic and heteroscedastic models,we propose a combination method based on OLS and GLS model averaging esti

会议

NEYMAN-PEARSON CLASSIFICATION UNDER HIGH-DIMENSIONAL SETTINGS

其他学术论文