建立多元混合回归模型分析影响毛兔产量的因素

本文把数量因子和非数量因子混合在一起，对５３５只长毛兔的实测数据进行了统计分析，建立了长毛兔产毛量的多元混合回归模型。并根据该模型分析了营养，兔舒气温等数量因子和作为非数量因子的５个生理阶段等因素对长毛兔产毛量的影响。     

混合效应模型及其在非寿险费率厘定中的应用

在非寿险分类费率厘定中,广义线性模型的应用十分普遍,但当某些费率因子的水平数很多时(本文称之为多水平因子),广义线性模型的估计结果将不可靠。解决此类问题的一种方法是把多水平费率因子作为随机效应处理。将多水平费率因子作为随机效应处理可以采取下述三种方法:(1)分别用广义线性模型和信度模型估计普通费率因子和多水平因子,通过广义线性模型与Buhlmann-Straub信度模型的迭代应用预测索赔频率和索赔强度;(2)应用广义线性混合模型分别预测索赔频率和索赔强度;(3)直接对经验纯保费数据建立Tweedie混合效应模型。本文把上述模型应用于中国车损险实际数据的研究结果表明,这三种方法比较接近,但从总体上看,广义线性混合模型的估计结果更加可取。     

半参数线性混合效应模型的联合变量选择

多数基于线性混合效应模型的变量选择方法分阶段对固定效应和随机效应进行选择,方法繁琐、易产生模型偏差,且大部分非参数和半参数的线性混合效应模型只涉及非参数部分的光滑度或者固定效应的选择,并未涉及非参变量或随机效应的选择。本文用B样条函数逼近非参数函数部分,从而把半参数线性混合效应模型转化为带逼近误差的线性混合效应模型。对随机效应的协方差矩阵采用改进的乔里斯基分解并重新参数化线性混合效应模型,接着对该模型的极大似然函数施加集群ALASSO惩罚和ALASSO惩罚两类惩罚,该法能实现非参数变量、固定效应和随机效应的联合变量选择,基于该法得出的估计量也满足相合性、稀疏性和Oracle性质。文章最后做了个数值模拟,模拟结果表明,本文提出的估计方法在变量选择的准确性、参数估计的精度两个方面均表现较好。     

联贷联保信用风险集中度定量分析北大核心CSCD

利用约化方法实现贷款组合渐进单因子模型,得到渐进单因子约化模型.为刻画联贷联保业务中联保体内的违约传染,又构建渐进单因子传染模型,使联保体内的违约相关由因子依赖和交叉强度共同体现,联保体间的违约相关只由因子依赖体现.通过模型求解得到联贷联保组合以及对比组合的违约概率,预期损失,非预期损失和集中度调整.数值分析比较了不同市场环境下联贷联保的风险,并据此建议商业银行限制担保链条长度和担保体内放贷额度,增加联保体数量,缩短贷款期限,在市场恶化时慎用联贷联保.     

采用拟合偏差分析法建立生长毛兔日增重模型

本文讨论了在建立生长毛兔的日增重──日粮蛋能比模型过程中，对具有重复测量的观测值，即对在每个点ｘｉ（ｉ＝ １， ２， ３，…ｋ）都对应有两个以上观测值 ｙｉｊ（ｊ＝２， ３， ４…，ｎ）的数据所假定的一元线性回归模型进行拟合偏差分析，检验线性模型的合理性，最终确定模型形式的方法．     

Quantitative analysis of concentration risk of joint loan business

利用约化方法实现贷款组合渐进单因子模型,得到渐进单因子约化模型.为刻画联贷联保业务中联保体内的违约传染,又构建渐进单因子传染模型,使联保体内的违约相关由因子依赖和交叉强度共同体现,联保体间的违约相关只由因子依赖体现.通过模型求解得到联贷联保组合以及对比组合的违约概率,预期损失,非预期损失和集中度调整.数值分析比较了不同市场环境下联贷联保的风险,并据此建议商业银行限制担保链条长度和担保体内放贷额度,增加联保体数量,缩短贷款期限,在市场恶化时慎用联贷联保.     

非参数混合效应模型中的删除诊断

本文考虑了非参数混合模型中的子集删除诊断．在非参数混合模型的框架下,得到了用两种方法所得非参数元素估计的系数矩阵相等,同时用两种方法得到的随机效应预测的系数矩阵也相同．在此等价性的基础上,我们将现有文献中对线性模型的11删除=替代”准则推广至非参数混合模型场合．不仅得到了固定效应的删除诊断公式,还得到了随机效应的删除诊断公式．此外,我们通过模拟数据分析和真实数据分析说明了我们方法的有效性．     

基于混合高斯模型的指标优化方法

借鉴混合高斯模型的方法,提出了基于混合高斯模型的非参数检验法用于评价指标体系的构建.首先运用混合高斯模型对现有指标体系进行聚类分析,然后运用非参数检验的秩相关系数对分类后的指标体系进行筛选,最终确定评价指标体系的构成,保证指标体系的全面性和代表性.实例结果表明,方法分类精度高、简便、实用,在指标优化中具有应用价值.     

偏正态数据下众数混合专家回归模型的参数估计

混合专家模型是对异质总体数据进行回归、分类和聚类的异构性建模的流行框架.研究基于偏正态分布,提出了众数混合专家回归模型,该模型既对混合偏态数据分类后进行众数建模,同时又对混合比例建模,相比单纯的众数回归模型具有更大的适应性,可以概括和描述众多的实际问题.采用了一种有效的模式识别聚类方法来选择子聚类的数量.分别应用MM算...     

共同因子结构下非平稳面板数据检验的一致性研究

为了避免伪检验结果,本文分析了多种共同因子结构下非平稳面板数据检验的一致性问题,发现相抵共同因子时面板单位根检验与协整检验存在不一致,进而构建了因子互冲面板协整模型,从模型分析、参数估计和假设检验等方面提出解决一致性问题的方法步骤,最后通过实证研究论证了因子互冲面板协整模型的可行性和实用性。     

关于非线性波动方程的爆破现象

通过引入一类“爆破因子K(u,ut）”,讨论了非线性波动方程分别具Newmann边界条件和Dirichlet边界条件时,混合问题对于常见的各种非线性情形及初值条件,解在有限时间内的爆破行为。     

Modeling the Oscillations of the CO + O2 Reaction Rate in a Granular Catalyst Layer

A new point model is proposed for the oxidation reaction of CO on the surface of a Pd cluster. The model reflects the oxidation–reduction mechanism on the palladium surface and incorporates all the stages of the kinetic scheme from the previous three-component model. A new assumption based on a series of experimental facts claims that the adsorbed oxygen atoms may diffuse into deeper-lying layers of the crystal lattice and thus influence the processes in the adsorption layer due to the micro size of the palladium clusters. The new point system has been built into the distributed general model for a granular catalyst developed previously. As a result, for parameter values corresponding to experimental conditions we have managed for the first time to obtain a wide region of chaos and complex mixed modes, close to those observed in real experiments.     

Ambiguity aversion and an intertemporal equilibrium model of catastrophe-linked securities pricing

To explain several stylized facts concerning catastrophe-linked securities premium spread, we propose an intertemporal equilibrium model by allowing agents to act in a robust control framework against model misspecification with respect to rare events. We have presented closed-form pricing formulas in some special cases and tested the model using empirical data and simulation.     

On an inequality of Minkowski for mixed volumes

Using a recently established stability result regarding the Brunn-Minkowski theorem and simple facts about convex functions we find strengthened versions of known inequalities for the mixed volumes of convex bodies. These results improve previously known inequalities of this type.Supported by National Science Foundation Research Grant DMS 870189.     

混合分数布朗运动环境下欧式期权定价

假设股票价格变化过程服从混合分数布朗运动,建立了混合分数布朗环境下支付连续红利的欧式股票期权的定价模型.利用混合分数布朗运动的It-公式,将支付连续红利的欧式股票期权的定价问题转化为一个偏微分方程,通过偏微分方程求解获得了混合分数布朗运动环境下支付连续红利的欧式股票看涨期权的定价公式.     

Optimal Mixed Batch Shipment Policy with Variable Safety Factor for the Single-Vendor Single-Buyer Production-Inventory System

This study investigates the production and inventory problem involving stochastic demand in a single-vendor single-buyer integrated system. The stochastic model is constructed and is controlled by both the reorder and shipping points with mixed geometric and equal batch shipment policy and variable safety factor. The minimum cost model is transformed into the maximum order quantity model. Consequently, the structure of mixed geometric and equal shipments can easily be obtained. The structure of the mixed geometric and equal shipments comprises number of batches, number of geometric shipments, and ratio of equal sized shipments to the size of the first shipment. The problem is solved using the proposed algorithm that determines the economic lot size, the structure of the mixed geometric and equal shipments, and the safety factor. An illustrative example demonstrates the effectiveness of the algorithmic procedures.     

Large eddy simulations of the turbulent flow between a rotating and a stationary disk

Large eddy simulations of the flow between a rotating and a stationary disk have been performed using a dynamic and a mixed dynamic subgrid-scale model. The simulations were compared to direct numerical simulation results. The mixed dynamic model gave better overall predictions than the dynamic model. Modifications of the near-wall structures caused by the mean flow three-dimensionality were also investigated. Conditional averages near strong stress-producing events led to the same conclusions regarding these modifications as studies of the flow generated by direct numerical simulation, namely a distinct asymmetry of the vortices producing sweeps and ejections.     

Large eddy simulations of the turbulent flow between a rotating and a stationary disk

Large eddy simulations of the flow between a rotating and a stationary disk have been performed using a dynamic and a mixed dynamic subgrid-scale model. The simulations were compared to direct numerical simulation results. The mixed dynamic model gave better overall predictions than the dynamic model. Modifications of the near-wall structures caused by the mean flow three-dimensionality were also investigated. Conditional averages near strong stress-producing events led to the same conclusions regarding these modifications as studies of the flow generated by direct numerical simulation, namely a distinct asymmetry of the vortices producing sweeps and ejections.     

A proof of the arithmetic-geometric mean inequality using non-Euclidean geometry

This note presents a proof of the arithmetic-geometric mean inequality that uses basic facts about the upper half-plane model of hyperbolic plane geometry. This material could find use as enrichment material in any model-oriented course on the classical geometries.     

国际油价波动跳跃性特征的实证分析

根据国际油价波动中存在异常跳跃的情况,本文运用EGARCH-Jump模型对国际油价波动的跳跃性特征进行了实证分析。结果表明,加入跳跃因素的模型减缓了国际油价波动的持续性,同时杠杆效应消失,表明跳跃性因素是国际油价波动的影响因素之一,也证实了国际油价波动的跳跃性特征是国际石油市场产生杠杆效应的原因。但从长期来看,跳跃性因素对国际油价波动的扰动影响并不大,国际油价的波动仍主要受正常信息的影响。总体上,EGARCH-Jump模型比普通GARCH族模型能更好地捕捉国际油价波动的动态性特征。