位置：成果数据库 > 期刊 > 期刊详情页
多维线性回归有偏子模型的多步调整相合推断
  • ISSN号：1003-3998
  • 期刊名称：数学物理学报
  • 时间：2012
  • 页码：1019-1031
  • 分类：O211[理学—概率论与数理统计;理学—数学]
  • 作者机构：[1]山东大学数学学院,济南250100, [2]山东省计算中心,济南250014, [3]山东师范大学数学科学学院,济南250014
  • 相关基金：国家自然科学基金（10921101,11171188）; 山东省自然科学基金（ZR2010AZ001,ZR2011AQ007）资助
  • 相关项目：非稀疏高维模型的重建和相合统计推断的研究
作者： 曾云辉|林路|王秀丽|
关键词： 线性回归, 有偏子模型, 相合估计, 主成分回归, 独立成分分析, Linear regression, Biased sub-model, Consistent estimation, Principal component regression, Independent component analysis
中文摘要：

当协变量是高维时经常采用一个子模型作为工作模型.由于没有包含所有相关的变量,这个模型可能是有偏的.这样,基于子模型得到的参数估计可能是不相合的.在这篇文章中将首先通过多步调整方法构造一个条件无偏模型.与现有的方法相比,这个调整模型仅采用了一维非参估计.然后得到子模型参数的一个全局相合估计,而且获得了该估计的渐近正态性.数值模拟结果显示,基于调整模型的参数估计优于基于子模型和全模型的参数估计.

英文摘要：

When the dimension of covariate is high,one usually uses a sub-model as working model.Such a model may be biased because not all relevant variables are contained in it.Thus the resulting estimator of parameter in the sub-model may be inconsistent.In this paper,we shall construct a conditionally unbiased model by multi-step-adjustment.Compared with the existing methods,the adjusted model only adopts univariate nonparametric estimations.A globally consistent estimator of parameter in the sub-model is constructed,and its asymptotic normality is also obtained.The simulation results further illustrate that the performance of the estimator based on the adjusted model is better than those of the estimators derived from the sub-model and the full model.

同期刊论文项目
非稀疏高维模型的重建和相合统计推断的研究
期刊论文 50
金融数学-金融风险控制中的G-风险度量、倒向随机分析与计算
期刊论文 56 会议论文 9
同项目期刊论文
Variable selection and parameter estimation for partially linear models via Dantzig selector
Block Empirical Likelihood for Longitudinal Single-Index Varying-Coefficient Model
协变量缺失下线性模型中参数的经验似然推断
Local linear–additive estimation for multiple nonparametric regressions
The dual and degrees of freedom of linearly constrainedgeneralized lasso.
The Dantzig Discriminant Analysis with High Dimensional Data
GMM and misspecification correction for misspecified models with diverging number of parameters
Simulation-based consistent inference for biased working model of non-sparse high-dimensional linear
The moment of maximum normed randomly weighted sums of martingale differences
Simultaneous structure estimation and variable selection in partial linear varying coefficient model
Walsh-average based variable selection for varying coefficient models
Variable Selection in Semiparametric Quantile Modeling for Longitudinal Data
New efficient estimation and variable selection in models with single-index structure
Derivative Estimation Based on Difference Sequence via Locally Weighted Least Squares Regression
Remodeling and Estimation for Sparse Partially Linear Regression Models
On general minimum lower order confounding criterion for s-level regular designs
k-sample upper expectation linear regression—Modeling, identifiability, estimation and prediction
变量约束工具变量回归及其在期权定价和投资组合中的应用
Independent feature screening for ultrahigh-dimensional models with interactions
2(n1+n2)？(k1+k2)42 混合水平裂区设计包含纯净效应的条件
Empirical likelihood analysis of longitudinal data involving within-subject correlation
Empirical likelihood for a varying coefficient partially linear model with diverging number of param
A stable multistep scheme for solving BSDEs
Backward stochastic viability and related properties on Z for BSDEs with applications
-error estimates for numerical schemes for solving certain kinds of backward stochastic differential
Dual method for continuous-time Markowitz’s problems with nonlinear wealth equations
Reflected Forward-Backward Stochastic Differential Equations with Continuous monotone Coefficients
Maximum principle for the stochastic optimal control problem with delay and application
Optimal Premium Policy of an Insurance Firm:
Maximum Principle for Partially- Observed Optimal Control of Fully- Coupled Forward-Backward Stochas
Finite element approximations for stokes-darcy flow with beavers-joseph interface conditions
On Jensen’s inequality and Holder’s inequality for g-expectation
BSDEs with random default time and related zero-sum
A Stochastic Maximum Principle for Optimal Control of Jump Diffusions and Applications to Finance
带反射边界的倒向随机微分方程解的收敛性
Relationship between MP and DPP for the Stochastic Optimal
Numerical solutions for forward backward doubly stochastic differential equations and zakai equation
A Maximum Principle for Partially Observed Optimal Control of Forward-Backward Stochastic Control Sy
Adaptive global confidence band for nonparametric
Backward stochastic differential equations with a uniformly continuous
A result on the uniqueness of generators of backward stochastic differential equations
The power of detecting enriched patterns:An HMM approach
A generalized Neyman–Pearson lemma under g-probabilitie
A Risk-Sensitive Stochastic Maximum Principle for Optimal Control of Jump Diffusions and its Applica
Function spaces and capacity related to a sublinear expectation: application to G-Brownian motion pa
BDSDEs with Locally Monotone Coefficients and Sobolev
Numerical algorithms for backward stochastic differential equations with 1-D Brownian motion: conver
On Model Selection Consistency of Bayesian Method for Normal Linear Models
Stochastic Maximum Principle for Optimal Control Problems of Forward -Backward Systems Involving Imp
A Comparison Theorem and Uniqueness Theorem of Backward Doubly Stochastic Differential Equations
容度的Borel-Cantelli引理
一类具有一致连续系数的倒向重随机微分方程
MAXIMUM PRINCIPLE FOR FORWARD-BACKWARD STOCHASTIC CONTROL SYSTEM WITH RANDOM JUMPS AND APPLICATIONS TO FINANCE
期刊信息
  • 《数学物理学报：A辑》
  • 北大核心期刊（2011版）
  • 主管单位:中国科学院
  • 主办单位:中国科学院武汉物理与数学研究所
  • 主编：李邦河 陈贵强 朱熹平
  • 地址：湖北省武汉市武昌小洪山西路30号武汉71010信箱
  • 邮编：430071
  • 邮箱：actams@wipm.ac.cn
  • 电话：027-87199206
  • 国际标准刊号：ISSN：1003-3998
  • 国内统一刊号：ISSN：42-1226/O
  • 邮发代号:38-214
  • 获奖情况:
  • 国内外数据库收录:
  • 俄罗斯文摘杂志,美国数学评论（网络版）,德国数学文摘,日本日本科学技术振兴机构数据库,中国中国科技核心期刊,中国北大核心期刊（2004版）,中国北大核心期刊（2008版）,中国北大核心期刊（2011版）,中国北大核心期刊（2014版）,中国北大核心期刊（2000版）
  • 被引量:5382