分数填补下两总体分位数差异的经验似然置信区间

设有两个非参数总体,其样本数据不完全,用分数填补法补足缺失数据,得到两总体的"完全"样本数据,在此基础上构造两总体分位数差异的经验似然置信区间.模拟结果显示,分数填补法可以得到更加精确的置信区间.     

分数填补下两非参数总体一般差异的经验似然置信区间

在一定的条件下证明了缺失数据情形基于分数填补方法得到的两非参数总体一般差异指标的经验似然比统计量的渐近分布为加权χ21,由此可构造差异指标的经验似然置信区间.     

响应变量缺失时条件分位数的经验似然置信区间

本文利用经验似然的思想,分别构造在响应变量满足随机缺失(MAR)机制的条件下,不含附加信息和含附加信息时条件分位数的置信区间,并说明检验的渐近功效随信息量的增加而非降,推广了现有文献中的相应结果.     

随机插补下两线性模型中响应变量分位数差异的经验似然置信区间

设两个样本数据不完全的线性模型,其中协变量的观测值不缺失,响应变量的观测值随机缺失。采用随机回归插补法对响应变量的缺失值进行补足,得到两个线性回归模型的"完全"样本数据,在一定条件下得到两响应变量分位数差异的对数经验似然比统计量的极限分布为加权x_1~2,并利用此结果构造分位数差异的经验似然置信区间。模拟结果表明在随机插补下得到的置信区间具有较高的覆盖精度。     

混合缺失机制下两总体差异的半经验似然置信区间

假定两个总体x与y均有数据缺失,它们的分布函数分别为F(·)与G_θ(·),其中F(·)未知,G_θ(·)的概率密度函数g_θ(·)形式已知,仅依赖于一些未知的参数,利用Fractional填补法填补缺失值,在一定的条件下证明了缺失数据下两总体差异指标的半经验似然比统计量的渐近分布为x_1~2,由此可构造两总体差异指标的经验似然置信区间.     

条件分位数和条件密度的经验似然置信区间

本文首次把经验似然引入非参数回归模型,分别得到了条件分位数和条件密度的经验似然置信区间。     

条件分位数的经验似然置信区间

本文利用经验似然思想分别讨论了不含附加信息和含附加信息时条件分位数的置信区间的构造,并考虑了一类检验问题,证明了检验的渐近功效随信息量的增加而非降．     

条件分位数的经验似然置信区间

本文利用经验似然思想分别讨论了不含附加信息和含附加信息时条件分位数的置信区间的构造,并考虑了一类检验问题,证明了检验的渐近功效随信息量的增加而非降.     

两样本分位数差异的半经验似然比检验

设x1，…，xn；y1，…，ym为独立随机样本，x1，…，xn同分布，x1～F（x），F未知，y1，…；ym同分布，y1～Gθ（y），Gθ（y）的形式已知，θ为未知参数．本文结合非参数似然思想和参数似然方法讨论F和Gθ的分位数差异的检验问题，在一定的条件下得到了半经验似然比统计量的渐近分布．     

具有随机右删失随机变量分位数的经验似然推断

本文研究了具有随机右删失随机变量分位数的置信域的构造.利用经验似然和截尾值估算相结合的方法,给出了分位数的对数经验似然比统计量,在较少的条件下证明了该统计量的极限分布为自由度为1的x~2分布.使得完全数据下的分位数的经验似然推断方法应用到非完全数据中.     

Semi-empirical likelihood confidence intervals for the differences of quantiles with missing data

Detecting population （group） differences is useful in many applications, such as medical research. In this paper, we explore the probabilistic theory for identifying the quantile differences .between two populations. Suppose that there are two populations x and y with missing data on both of them, where x is nonparametric and y is parametric. We are interested in constructing confidence intervals on the quantile differences of x and y. Random hot deck imputation is used to fill in missing data. Semi-empirical likelihood confidence intervals on the differences are constructed.     

Empirical likelihood method for quantiles with response data missing at random

Empirical likelihood is a nonparametric method for constructing confidence intervals and tests,notably in enabling the shape of a confidence region determined by the sample data.This paper presents a new version of the empirical likelihood method for quantiles under kernel regression imputation to adapt missing response data.It eliminates the need to solve nonlinear equations,and it is easy to apply.We also consider exponential empirical likelihood as an alternative method.Numerical results are presented to compare our method with others.     

Empirical likelihood confidence intervals for the differences of quantiles with missing data

Suppose that there are two nonparametric populations x and y with missing data on both of them. We are interested in constructing confidence intervals on the quantile differences of x and y. Random imputation is used. Empirical likelihood confidence intervals on the differences are constructed. Supported by the National Natural Science Foundation of China (No. 10661003) and Natural Science Foundation of Guangxi (No. 0728092).     

Empilrical likelihood for non-parametric regression models with missing responses: Multiple design case

Empirical likelihood(EL) ratio statistic on θ = g(x) is constructed based on the inverse probability weighted imputation approach in a nonparametric regression model Y = g(x) + ε(x ∈ [0,1]p) with fixed designs and missing responses,which asymptotically has χ12 distribution.This result is used to obtain a EL based confidence interval on θ.     

Semi-empiricial Likelihood Confidence Intervals for the Differences of Two Populations Based on Fractional Imputation

Suppose that there are two populations x and y with missing data on both of them, where x has a distribution function F（·） which is unknown and y has a distribution function Gθ（·） with a probability density function gθ（·） with known form depending on some unknown parameter θ. Fractional imputation is used to fill in missing data. The asymptotic distributions of the semi-empirical likelihood ration statistic are obtained under some mild conditions. Then, empirical likelihood confidence intervals on the differences of x and y are constructed.     

Adjusted empirical likelihood method for quantiles

Empirical likelihood (EL) was first applied to quantiles by Chen and Hall (1993,Ann. Statist.,21, 1166–1181). In this paper, we shall propose an alternative EL approach which is also some kind of the kernel method. It not only eliminates the need to solve nonlinear equations, but also is extremely easy to implement. Confidence intervals derived from the proposed approach are shown, by an nonparametric version of Wilks' theorem, to have the same order of coverage accuracy (order 1/n) as those of Chen and Hall. Numerical results are presented to compare our method with other methods.     

响应变量存在缺失时部分线性模型的经验似然推断

考虑响应变量带有缺失的部分线性模型,采用借补的思想,研究了参数部分和非参数部分的经验似然推断,证明了所提出的经验对数似然比统计量依分布收敛到χ2分布,由此构造参数部分和函数部分的置信域和逐点置信区间.对参数部分,模拟比较了经验似然与正态逼近方法;对函数部分,模拟了函数的逐点置信区间.