Sparse Supervised Dimension Reduction in High Dimensional Classification

来源 :中国数量经济学会 | 被引量 : 0次 | 上传用户：21stsun

【摘要】

：

【作者】

：

Junhui Wang

【机构】

：

Department of Mathematics, Statistics,and Computer Science University of Illinois at Chicago

【出处】

：

中国数量经济学会

【发表日期】

：

2009年12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

　　Supervised dimension reduction has proven to be effective in analyzing data with complex structure.The primary goal is to seek the reduced subspace of minimal dimension which is sufficient for summarizing the data structure of interest.In this talk,I will discuss the sufficient dimension reduction in classification context,and proposes a novel method which is capable of estimating the dimension reduction subspace while retaining the ideal classification boundary based on the original dataset.

其他文献

An Empirical Investigation of the ACD Model for Trading Price:Comparison and Selection

This paper compares the dynamic fitting behavior of parametric and nonparametric trading duration models,and selects an appropriate autoregressive conditional duration(ACD) model for analyzing price d

会议

ACD modelirregularly spaced datanonparametric estimationMonte Carlo simulatio

Variable selection with multiple heterogeneous responses

Consider a heterogeneous population composed of multiple sub-groups.The responses in different groups may share common "important" covariates,while having group-specific important covariates.With such

会议

Stochastic Dominance via Quantile Regression

We derive a new way to test for stochastic dominance between the return of two assets using a quantile regression formulation.The test statistic is a variant of the one-sided Kolmogorov-Smirno statist

会议

Factors Influencing International Expansion of US-Based Fast Food Franchise Systems

This study examines the factors on the decision whether to seek overseas expansion by US-based fast food franchise systems.It develops a theoretical framework mainly based on the agency theory to iden

会议

Nonparametric Inferences for the Lorenz Curve and the Generalized Lorenz Curve

In this talk,we discuss the empirical likelihood-based inferences for the Lorenz curve (LC) and the Generalized Lorenz curve (GLC).In the settings of simple random sampling,stratified random sampling

会议

Mutual Funds Evaluation with Data Mining Method

There are many Mutual Funds rating companies in the market.Rating companies have different standards to evaluate Mutual Funds such as risks,incomes and some referable indexes.The study is aim at setti

会议

Joint Regression Analysis of Correlated Data Using Gaussian Copulas

We present a new joint modeling approach for correlated data analysis.Utilizing Gaussian copulas,we develop a unified and flexible machinery to integrate separate one-dimensional generalized linear mo

会议

Efficient estimation of stratified mark-specific proportional hazards models withmissing marks

An objective of randomized placebo-controlled preventive HIV vaccine efficacy trials is to assess the relationship between the vaccine effect to prevent infection and the genetic distance of the expos

会议

Data sample selection issues for bankruptcy prediction

Bankruptcy prediction is of paramount interest to both academics and practitioners.This paper devotes special care to an important aspect of the bankruptcy prediction modeling:data sample selection is

会议

Bankruptcy ForecastingBinary ClassificationChoice Based Data

Asymptotic Property of Fixed-Design Semiparametric RegressionModel with Nonparametric AR(1)Errors

Under fixed-design,considered semiparametric regression model with the non-parametric AR (1) errors.This paper integrate least squares estimation method and nonparametric kernel estimation method,defi

会议

Semiparametric regression modelFixed-designNon-parametric AR(1) errors

Sparse Supervised Dimension Reduction in High Dimensional Classification

与本文相关的学术论文