Regularized multivariate analysis
Regularized multivariate analysis
批准号:
6394-2006
负责人:
Takane, Yoshio
金额:
$1.46万
依托单位:
依托单位国家:
加拿大
项目类别:
Discovery Grants Program - Individual
财政年份:
2006
资助国家:
加拿大
项目状态:
已结题
起止时间:
2006-01-01 至 2007-12-31
中文摘要
到目前为止,统计学理论在很大程度上被渐近理论所主导。然而,我们经常不得不处理小样本的数据。在这种情况下,最大似然估计(MLE)被证明是对大样本最优的,但可能不是最优的。我建议在这种情况下使用正则化估计。正则化是指在数据分析中利用先验知识的任何过程,包括真正被称为惩罚、平滑、收缩、软约束等的所有过程。更具体地说,我建议在冗余分析(降阶回归)、多集典型相关分析以及其他多变量分析(MVA)技术中加入脊型以及其他类型的收缩估计。众所周知,这种类型的估计提供与较小的均方误差(真实参数值和估计之间的平均差异)相关的参数估计,特别是在小样本中。我们使用实际数据和模拟方法证明了正则化(收缩估计)在这些技术中的有效性。我们计划编写一系列用户友好的计算机程序并将其公之于众,使正则化的MVA成为多元数据分析的常规部分。
英文摘要
The theory of statistics has so far largely been dominated by asymptotic theories. However, we often have to deal with small sample sized data. In such situations, the maximum likelihood estimation (MLE), proven optimal for large samples, may not be optimal. I propose to use regularized estimation in such situations. The regularization means any processes that utilize prior knowledge in data analysis, and includes all such processes that are veriously called penalizing, smoothing, shrinking, soft constraining, etc. More specifically, I propose to incorporate a ridge type as well as other types of shrinkage estimation in redundancy analysis (reduced rank regression), multiple-set canonical correlation analysis as well as other techniques of multivariate analysis (MVA). This type of estimation is known to provide parameter estimates associated with smaller mean square errors (average discrepancies between true parameter values and the estimates) particularly in small samples. We demonstrate the effectiveness of regularization (shrinkage estimation) in these techniques, using both real data and simulation methods. We plan to write a series of user friendly computer programs and make them publicly available to make regularized MVA a routine part of multivariate data analysis.
期刊论文(0)
专著(0)
科研奖励(0)
会议论文
Developments of techniques for multivariate data analysis
-
批准号:6394-2009
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$1.17万
-
财政年份:2015
-
负责人:Takane, Yoshio
-
依托单位:
Developments of techniques for multivariate data analysis
-
批准号:6394-2009
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$1.17万
-
财政年份:2012
-
负责人:Takane, Yoshio
-
依托单位:
Developments of techniques for multivariate data analysis
-
批准号:6394-2009
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$1.17万
-
财政年份:2011
-
负责人:Takane, Yoshio
-
依托单位:
Developments of techniques for multivariate data analysis
-
批准号:6394-2009
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$1.17万
-
财政年份:2010
-
负责人:Takane, Yoshio
-
依托单位:
Developments of techniques for multivariate data analysis
-
批准号:6394-2009
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$1.17万
-
财政年份:2009
-
负责人:Takane, Yoshio
-
依托单位:
Regularized multivariate analysis
-
批准号:6394-2006
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$1.46万
-
财政年份:2008
-
负责人:Takane, Yoshio
-
依托单位:
Regularized multivariate analysis
-
批准号:6394-2006
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$1.46万
-
财政年份:2007
-
负责人:Takane, Yoshio
-
依托单位:
Structured analysis of multivariate data in psychology
-
批准号:6394-2001
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$2.19万
-
财政年份:2005
-
负责人:Takane, Yoshio
-
依托单位:
Structured analysis of multivariate data in psychology
-
批准号:6394-2001
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$2.19万
-
财政年份:2004
-
负责人:Takane, Yoshio
-
依托单位:
Structured analysis of multivariate data in psychology
-
批准号:6394-2001
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$2.19万
-
财政年份:2003
-
负责人:Takane, Yoshio
-
依托单位:
Structured analysis of multivariate data in psychology
-
批准号:6394-2001
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$2.19万
-
财政年份:2002
-
负责人:Takane, Yoshio
-
依托单位:
Structured analysis of multivariate data in psychology
-
批准号:6394-2001
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$2.19万
-
财政年份:2001
-
负责人:Takane, Yoshio
-
依托单位:
Extensions of constrained canonical correlation analysis and develop- ment of neural network models for missing data problems
-
批准号:6394-1997
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$2.19万
-
财政年份:2000
-
负责人:Takane, Yoshio
-
依托单位:
Extensions of constrained canonical correlation analysis and develop- ment of neural network models for missing data problems
-
批准号:6394-1997
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$2.19万
-
财政年份:1999
-
负责人:Takane, Yoshio
-
依托单位:
Extensions of constrained canonical correlation analysis and develop- ment of neural network models for missing data problems
-
批准号:6394-1997
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$2.08万
-
财政年份:1998
-
负责人:Takane, Yoshio
-
依托单位:
Extensions of constrained canonical correlation analysis and develop- ment of neural network models for missing data problems
-
批准号:6394-1997
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$1.89万
-
财政年份:1997
-
负责人:Takane, Yoshio
-
依托单位:
Extensions of constrained principal component analysis and other psychometric models
-
批准号:6394-1993
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$2.04万
-
财政年份:1996
-
负责人:Takane, Yoshio
-
依托单位:
Extensions of constrained principal component analysis and other psychometric models
-
批准号:6394-1993
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$2.04万
-
财政年份:1995
-
负责人:Takane, Yoshio
-
依托单位:
Extensions of constrained principal component analysis and other psychometric models
-
批准号:6394-1993
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$2.04万
-
财政年份:1994
-
负责人:Takane, Yoshio
-
依托单位:
Extensions of constrained principal component analysis and other psychometric models
-
批准号:6394-1993
-
项目类别:Discovery Grants Program - Individual
-
资助金额:$2.04万
-
财政年份:1993
-
负责人:Takane, Yoshio
-
依托单位:
国内基金
海外基金
基于线性及非线性模型的高维金融时间序列建模:理论及应用
-
批准号:71771224
-
项目类别:面上项目
-
资助金额:49.0万元
-
批准年份:2017
-
负责人:王辉
-
依托单位: