2. 您的位置
3. 首页
4. 会议论文
5. Maximum likelihood inference for a large precision matrix

Maximum likelihood inference for a large precision matrix

来源 :上海交通大学 | 被引量 : 0次 | 上传用户：vertra

【摘 要】

：

Inference concerning Gaussian graphical models involves pairwise conditional dependencies on Gaus-sian random variables.In such a situation,regularization of a certain form is often employed to treat

【作 者】

：

Yunzhang Zhu Xiaotong Shen Wei Pan 

【机 构】

：

Department of Statistics,Ohio State University,43210,USA

【出 处】

：

上海交通大学

【发表日期】

：

2016年4期

【关键词】

：

Constrained likelihood inference Generalized Wilks phenomenon Graphical models H 

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

　　Inference concerning Gaussian graphical models involves pairwise conditional dependencies on Gaus-sian random variables.In such a situation,regularization of a certain form is often employed to treat an overparameterized model,imposing challenges to inference.

其他文献

MONITORING MULTIVARIATE PROCESSES WITH GROUPED CHANGES

We consider the multivariate process monitoring when a group of variables with strong correlation tends to suffer from abnormal changes simultaneously.Such a problem arises naturally in many practical

会议

False Discovery RateMultivariate Process ControlMultivariate EWMA chartGroupe

Bernstein Inequality for Multivariate Point Processes

At first we review some classical results on Bernstein type inequalities and latest progress.Then we introduce our new results on a Bernstein type concentration inequality for multivariate point proce

会议

Bernstein InequalityMultivariate Point Processgeneric chaining methodHellinge

Biased Random Walks on Groups and Graphs

In this talk,we consider a class of biased random walks on infinite groups and graphs,and study spectral radius,escape speed,heat kernel,and free spanning forest and wired spanning forest for these ra

会议

Random WalkInfinite GroupInfinite Graph

How to work beyond being a statistical copycat in China Medicines Development?

In China medicines development,factors,including medical practice,CMC,regulatory,and many others,often prevent researchers from joining Multi-Regional Clinical Trials.

会议

Lasso adjustments of treatment effect estimates in randomized experiments

We provide a principled way for investigators to analyze randomized experiments when the number of covariates is large.Investigators often use linear multivariate regression to analyze randomized expe

会议

randomized experimentNeyman-Rubin modelaverage treatment effecthigh-dimension

CLT for linear spectral statistics of large dimensional sample covariance matrices with dependent da

This paper investigates the central limit theorem for linear spectral statistics of high dimensional sample covariance matrices of the form Bn = n-1∑nj =1 Qxjx*j Q* where Q is a nonrandom matrix of di

会议

Large sample covariance matriceslinear spectral statisticscentral limit theore

Spatial Covariance Function with Stochastic Volatility

Geostatistical data has received great attention in recent years.To model spatial dependence of such type of data,the spatial covariance function is commonly employed.

会议

Spatial Covariance FunctionSpatial Missing DataStochastic VolatilitySpatial K

GRAPHICAL LATENT REGRESSION-MULTINOMIAL MIXTURE FOR MODELING COMPLEX DEPENDENCE

We have seen complex dependence in many real fields including biology,advertising,and real estate.Their data are complex and heterogeneous in general.

会议

Multinomial MixtureMarkov Random Field

ON THE INCREASING PATHS IN ACCESSIBILITY PERCOLATION

Consider a regular n-ary tree of height h,where n = αh.Each vertex is assigned a real number,called its fitness.Except that the fitness value of the root is always assumed to be 0,the fitness values a

会议

Increasing pathsaccessibility percolationregular tree

HYBRID GRAPHICAL LEAST SQUARE ESTIMATION AND ITS APPLICATION IN PORTFOLIO SELECTION

A novel regression method based on the idea of graphical models is proposed to deal with the portfolio optimisation problem within the Markowitz mean-variance framework,when the number of assets V is

会议

Graphical ModelGraphical Least SquaresLASSORidge RegressionUnbiased Estimati