Prediction risk for global-local shrinkage regression

来源 :上海交通大学 | 被引量 : 0次 | 上传用户：jiu999

【摘要】

：

Predictive performance in shrinkage regression suffers from two major difficulties:(ⅰ)the amount of relative shrinkage is monotone in the singular values of the design matrix and(ⅱ)the amount of shrin

【作者】

：

Anindya Bhadra

【出处】

：

上海交通大学

【发表日期】

：

2016年4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

其他文献

Nonparametric Robust Statistics and Outlier Reset CUSUM for the Exploration of Copy Number Alteratio

Copy number alteration(CNA)data have been collected to study disease related chro-mosomal amplifications and deletions.

会议

copy number alterationCUSUMoutlier

Adaptive multi-factor multi-stage clinical trial methods

Efficacious and effective interventions usually involve multiple factors.And it is fantasy to have pilot information about the main effects and their interactions for all factors at the time of initia

会议

factorial designclinical trialadaptivemulti-stage

Bias correction in subgroup analysis

Subgroup analysis becomes a popular practice in analyzing clinical trial data for identifying subgroup of patients who may benefit from the treatment of interest.

会议

bias correctionrandomized clinical trialsubgroup analysismultiple tests

Feature Selection in High-Dimensional Survival Data with Linear Regression

Variable selection from the high-dimensional survival data is a fundamental and challenging approach in recent years.

会议

Censored dataOracle propertyVariable selection

Stochastic Majorization-Minimization for Machine Learning

Stochastic optimization is a naturally attracting more attention in order to handle the growing size of data.

会议

MM AlgorithmStochastic OptimizationMachine Learning

Modelling asymmetric bivariate data using mixtures of rotated copulas

Copulas are widely used for modelling complex multivariate dependencies.Our study offers a framework for the application of pair-mixtures of rotated copulas to model asymmetric dependencies in bivaria

会议

Effect of Imputation of Missing Data in the State Inpatient Databases on Racial Disparities Research

Research Objective: Racial disparities(RD)in healthcare outcomes in the U.S.have been identified in recent decades for total joint arthroplasty,particularly total knee arthroplasty(TKA).

会议

Coupling Computer Models through Linking their Statistical Emulators

Gaussian processes,together with an objective Bayesian implementation of the processes,have become a common tool for emulating(approximating)complex computer models.

会议

Uncertainty QuantificationComputer modelsGaussian ProcessesBayesian statistic

Local Shrinkage Covariance Model

Estimation of covariance matrix function is fundamentally important for many problems.

会议

Local shrinkagecovariance matrix functionMEG neuroimaging

Fitting statistical models is computationally challenging when the sample size or the dimension of the dataset is huge.

会议

Distributed computingEmbarassingly parallelHigh dimensionalitySparse regressi

Prediction risk for global-local shrinkage regression

其他学术论文