缺失数据下线性模型中回归参数的经验似然置信区域

在缺失样本下,构造了线性模型中参数的调整的经验似然置信域,数值模拟表明调整的经验似然置信域有较好的覆盖率和精度.     

随机右截尾情形下两个指数分布总体的比较

研究了随机右截尾情形下两个指数分布的参数估计.在两个指数分布的刻度参数都未知时,得到了两个指数分布的刻度参数的最大似然估计与参数比的区间估计.     

A maximum principle for smooth optimal impulsive control problems with multipoint state constraints

A nonlinear optimal impulsive control problem with trajectories of bounded variation subject to intermediate state constraints at a finite number on nonfixed instants of time is considered. Features of this problem are discussed from the viewpoint of the extension of the classical optimal control problem with the corresponding state constraints. A necessary optimality condition is formulated in the form of a smooth maximum principle; thorough comments are given, a short proof is presented, and examples are discussed.     

核实数据下响应变量缺失的线性EV模型经验似然推断

考虑响应变量随机缺失而协变量带有误差的线性模型,借助于核实数据和借补方法,构造了回归系数的两种经验似然比,证明了所提出的估计的经验对数似然比渐近于一个自由度为1的独立χ2变量的加权和;而经调整后所得的调整经验对数似然比渐近于自由度为p的χ2分布,该结果可以用来构造未知参数的置信域.此外,我们也构造了响应均值的调整经验对数似然比统计量,并证明了所提出的统计量渐近于x2分布,可用此结果构造响应均值的置信域.通过模拟研究比较了置信域的精度及其平均区间长度.     

消费者不确定性的测度——基于异方差的视角

预防性储蓄理论的核心是对消费者不确定的测度.认为异方差是不确定性的很好代理变量.本文以极大似然法估计异方差,并运用消费者家计抽样调查数据,讨论收入和支出的异方差对消费者行为的影响.     

变系数混合模型的光滑样条推断

为了拟合纵向数据和其他相关数据,本文提出了变系数混合效应模型(VCMM).该模型运用变系数线性部分来表示协变量对响应变量的影响,而用随机效应来描述纵向数据组内的相关性, 因此,该模型允许协变量和响应变量之间存在十分灵活的泛函关系.文中运用光滑样条来估计均值部分的系数函数,而用限制最大似然的方法同时估计出光滑参数和方差成分,我们还得到了所提估计的计算方法.大量的模拟研究表明对于具有各种协方差结构的变系数混合效应模型,运用本文所提出的方法都能够十分有效地估计出模型中的系数函数和方差成分.     

具有线性趋势的回归信度模型中的估计和检验

研究具有线性趋势回归信度模型的参数估计和检验. 对该模型的回归系数和随机效应的方差,利用正交变换法得到了它们的极大似然估计, 并得到了参数的无偏估计. 对随机效应和是否有线性趋势采用似然比检验, 得到了似然统计量较好的近似$P$值, 并对检验的功效进行了模拟研究.     

正交饱和效应模型及其统计分析

文\cite{14}讨论了2水平正交饱和设计的一种统计分析方法---零效应搜索法, 本文将此方法推广到一般的饱和模型---正交饱和效应模型. 首先给出正交饱和效应模型的定义及其相关的参数估计和平方和分解, 然后用零效应搜索法对显著效应的检验问题进行分析, 模拟结果显示效果很好.     

煤浆浓度和颗粒分布对煤浆黏度预测的影响

采用神华煤制备煤浆,分析了颗粒粒径比λ和小颗粒体积分数ξ对双峰分布浆体黏度的影响,根据浆体黏度的关联式预测了煤浆的黏度并且与实验结果进行了比较。结果表明,采用双峰分布的颗粒制浆可以有效地降低浆体的黏度,同时可以获得较大的浆体体积分数 。在相同体积分数下,随着颗粒粒径比λ的增加,浆体的黏度迅速下降。当小颗粒体积分数ξ_dp1为35%时,浆体的黏度最小。采用Ouchiyama模型计算浆体的最大体积分数Φ_m与实验值较为吻合,而浆体的本质黏度[μ]基本保持不变。考虑λ和小颗粒体积分数ξ对双峰分布浆体的最大体积分数Φ_m的影响,可以采用单峰分布浆体的黏度关联式预测双峰分布浆体的黏度。     

Budget-constrained optimal insurance with belief heterogeneity

We re-examine the problem of budget-constrained demand for insurance indemnification when the insured and the insurer disagree about the likelihoods associated with the realizations of the insurable loss. For ease of comparison with the classical literature, we adopt the original setting of Arrow (1971), but allow for divergence in beliefs between the insurer and the insured; and in particular for singularity between these beliefs, that is, disagreement about zero-probability events. We do not impose the no sabotage condition on admissible indemnities. Instead, we impose a state-verification cost that the insurer can incur in order to verify the loss severity, which rules out ex post moral hazard issues that could otherwise arise from possible misreporting of the loss by the insured. Under a mild consistency requirement between these beliefs that is weaker than the Monotone Likelihood Ratio (MLR) condition, we characterize the optimal indemnity and show that it has a simple two-part structure: full insurance on an event to which the insurer assigns zero probability, and a variable deductible on the complement of this event, which depends on the state of the world through a likelihood ratio. The latter is obtained from a Lebesgue decomposition of the insured’s belief with respect to the insurer’s belief.