Computational Precision of the Power Function for Conditional Tests of Assumptions of the Rasch Model

Draxler and Zessin [1] derived the power function for a class of conditional tests of assumptions of a psychometric model known as the Rasch model and suggested an MCMC approach developed by Verhelst [2] for the numerical approximation of the power of the tests. In this contribution, the precision of the Verhelst approach is investigated and compared with an exact sampling procedure proposed by Miller and Harrison [3] for which the discrete probability distribution to be sampled from is exactly known. Results show no substantial differences between the two numerical procedures and quite accurate power computations. Regarding the question of computing time the Verhelst approach will have to be considered much more efficient.

The Survival Probability in Generalized Poisson Risk Model

In this paper we generalize the aggregated premium income process from a constant rate process to a poisson process for the classical compound Poinsson risk model,then for the generalized model and the classical compound poisson risk model ,we respectively get its survival probability in finite time period in case of exponential claim amounts.

Application of Brownian model in the north- western Beijing, China

The mathematic theory of Brownian passage-time model and its difference from other recurrence models such as Poisson, lognormal, gamma and Weibull, were introduced. We assessed and analyzed the earthquake probabilities of the major faults with the elapsed time much greater than the recurrence interval in the northwest region of Bei- jing (China) in 100-year by using both Brownian passage-time model and Poisson model, and concluded that the calculated results obtained from Brownian passage-time model is more reasonable.

Improved stress release model: Application to the study of earthquake prediction in Taiwan Region

Stress release model used to be applied to seismicity study of large historical earthquakes in a space of large scale. In this paper, we improve the stress release model, and discuss whether the stress release model is still applicable or not in the case of smaller spatio-temporal scale and weaker earthquakes. As an example of testing the model, we have analyzed the M greater than or equal 6 earthquakes in recent about 100 years. The result shows that the stress release model is still applicable. The earthquake conditional probability intensity in Taiwan Region is calculated with the improved stress release model. We see that accuracy of earthquake occurrence time predicted by the improved stress release model is higher than that by Poisson model in the test of retrospect earthquake prediction.

Study on the Recurrence Probability of Strong Earthquakes of Faults

Based on the physical model of Brownian passage time,the probabilities of recurrence of strong earthquakes on the major active faults in China are calculated in different predictive time spans,based mainly on the analysis of the earthquake preparation process before a strong earthquake occurs. Furthermore,the seismic risks on active faults are studied. The results show that the earthquake probabilities on the Xianshuihe fault,the Altyn Tagh fault,the east Kunlun fault and Xiaojiang fault are significantly greater than other faults in the Chinese mainland,which indicates that the level of stress accumulation on these faults are higher than on other faults. Therefore,these faults may have a seismic risk for strong earthquake in future.

ON THE BASIC REPRODUCTION NUMBER OF GENERAL BRANCHING PROCESSES

Under a very general condition （TNC condition） we show that the spectral radius of the kernel of a general branching process is a threshold parameter and hence plays a role as the basic reproduction number in usual CMJ processes. We discuss also some properties of the extinction probability and the generating operator of general branching processes. As an application in epidemics, in the final section we suggest a generalization of SIR model which can describe infectious diseases transmission in an inhomogeneous population.

Estimating Design Loads with Environmental Contour Approach Using Copulas for an Offshore Jacket Platform

Jacket-type offshore platforms are widely used for oil, gas field, and energy development in shallow water. The design of a jacket structure is highly dependent on target environmental variables. This study focuses on a strategy to estimate design loads for offshore jacket structures based on an environmental contour approach. In addition to the popular conditional distribution model, various classes of bivariate copulas are adopted to construct joint distributions of environmental variables. Analytical formulations of environmental contours based on various models are presented and discussed in this study. The design loads are examined by dynamic response analysis of jacket platform. Results suggest that the conditional model is not recommended for use in estimating design loads in sampling locations due to poor fitting results. Independent copula produces conservative design loads and the extreme response obtained using the conditional model are smaller than those determined by copulas. The suitability of a model for contour construction varies with the origin of wave data. This study provides a reference for the design load estimation of jacket structures and offers an alternative procedure to determine the design criteria for offshore structures.

On tail behavior of nonlinear autoregressive functional conditional heteroscedastic model with heavy-tailed innovations

We study the tail probability of the stationary distribution of nonparametric nonlinear autoregressive functional conditional heteroscedastic (NARFCH) model with heavytailed innovations. Our result shows that the tail of the stationary marginal distribution of an NARFCH series is heavily dependent on its conditional variance. When the innovations are heavy-tailed, the tail of the stationary marginal distribution of the series will become heavier or thinner than that of its innovations. We give some specific formulas to show how the increment or decrement of tail heaviness depends on the assumption on the conditional variance function. Some examples are given.

Long－ and intermediate－termseismic poten－tial of Fen－Wei seismic belt：active fault data application

Recurrence model for strong earthquakes on Fen Wei seismic belt is proposed on the basis of the collection and analysis of fault slip rate, paleoearthquake sequence, maximum displacement of each event etc. on 21 fault segments of the belt, which are active since late Late Pleistocene. And the long and intermediate term seismic potential of the belt has been evaluated through four approaches.

原发性肝癌患者生存期的危险因素及贝叶斯网络模型分析

目的探讨影响原发性肝癌(PHC)患者30个月生存期的危险因素及因素间复杂的网络关系。方法回顾性分析2016年10月至2020年11月山西医科大学第一医院363例PHC患者临床资料。通过随访了解患者的生存状况等,从最初诊断时开始计算,以月为单位,观察期截断点为2022年10月。根据文献报道及本研究中PHC患者的平均生存时间(29.6个月),最终将生存时间的研究节点定为30个月。采用多因素Logistic回归分析影响PHC患者30个月生存期的危险因素,并建立贝叶斯网络模型,以深入分析这些危险因素间复杂的网络关系,再通过受试者操作特征(ROC)曲线和校准度散点图(calibration plot)对模型的效能进行评价。结果363例PHC患者中生存时间<30个月的患者有167例(46.01%)。单因素分析结果显示,两组患者在γ-谷氨酰转移酶(γ-GT)水平、肿瘤大小、有无远处转移等方面差异有统计学意义(P<0.05);多因素logistic回归分析结果显示,γ-GT(>40 U/L)、肿瘤大小(>3 cm)、伴有远处转移均为PHC患者30个月生存期的危险因素(P<0.05);贝叶斯网络模型结果显示肿瘤大小、γ-GT水平和有无远处转移是直接影响PHC患者30个月生存期的危险因素,肿瘤个数和年龄则是间接危险因素;条件概率分布表显示当节点肿瘤大小>3 cm、γ-GT>40 U/L,伴有远处转移的患者,PHC生存时间<30个月的概率最大(98.408%)。结论肿瘤大小>3 cm、γ-GT>40 U/L和伴有远处转移是影响PHC患者30个月生存期的危险因素,年龄通过影响肿瘤大小和有无远处转移与PHC患者30个月生存期间接相关。

基于归一化流概率模型的水电机组异常声音检测

为推进水电机组智能化运维的发展,提出了一种自注意多阶统计量池化(SAMOSP)归一化流条件概率模型(NFCPM)用于水电机组的无监督异常声音检测。文中首次提出了自注意多阶统计量池化模块。该模块首先用一维压缩卷积层和瓶颈压缩激励部分自注意到时间帧的权重向量。权重向量用来计算多阶统计池化向量。然后再分频段的自注意到多阶统计池化量的不同权重,并按其提取不同频段的重要统计量信息,从而得到时频图的自注意统计池化特征向量。接着用归一化流条件概率模型对从水轮机组正常音频信号中提取到的自注意统计池化特征向量进行正常数据的概率建模。不同时间段的测试样本在该正常数据概率模型中进行测试,并得到相应的分数。分数越低表示概率密度越低,则异常程度越大,从而实现水电机组音频信号的无监督异常检测。

Logic, Pragmatics, and Types of Conditionals

Johnson-Laird and Byrne distinguished ten kinds of conditionals. Their framework was the mental models theory and they attributed different combinations of semantic possibilities to those ten types of conditionals. Based on such combinations, the mental models theory has clear predictions for reasoning tasks, including those kinds of conditionals and involving reasoning schemata such as Modus Ponens, Modus Tollens, the affirming the consequent fallacy, and the denying the antecedent fallacy. My aim in this paper is to show that the prediction,; of the mental logic theory for those reasoning tasks are exactly the same as those of the mental models theory, and that, therefore, such tasks are not useful to decide which of the two theories is correct.

潜在类别分析技术在心理学研究中的应用

潜在类别分析是通过对类别型的外显变量和潜在变量之间的关系建立统计模型,根据模型参数得到各种潜在类别的具体外在表现的潜在特征分类技术。该分析方法主要应用于心理行为特征的分类、控制认知心理实验中被试个体差异引起的系统误差、评价临床心理诊断的精确性,以及心理测验中的项目分析、信度分析、结构分析等。对此方法的优劣进行分析比较,表明:该方法可以与其他测量理论相结合进一步拓展其在心理测量中的应用,也可在纵向数据和多水平数据中应用。在应用中亦有提升方法技术的空间。

考虑父节点的贝叶斯网络故障路径追溯算法

基于概率图模型的故障诊断技术常被用于不确定信息的推理。其中贝叶斯网络作为一类具有强大实用性及适用性的概率图模型,得到了广泛应用。为有效识别大数据量工业系统中的故障传播路径,提出了一种考虑父节点影响的贝叶斯网络故障路径追溯算法。在建立贝叶斯网络后,给定父节点所有潜在取值组合,算法利用条件概率分解及二分法估计特定故障子节点的最大条件分布值,然后与真实观测值比较,得到导致此故障发生的最大可能原因。通过逐层推断,最终能够实现对网络中故障传播路径的有效追溯。实验结果表明,该算法具有准确追踪故障路径的能力,同时显示出较高的可用性及较快的追踪速度。

非一致性水文频率分析的研究进展

水文频率分析计算需满足独立随机同分布假设,其中同分布是指水文样本在过去、现在和未来均服从同一总体分布,即样本应具有一致性。然而,由于气候变化及人类活动的影响,使得一致性的假设受到挑战,因此变化环境下传统频率计算方法获得的设计结果,其可靠性受到质疑。为此,探索适应非一致性极值系列的频率分析方法已显得非常重要。总结介绍国内外关于非一致性水文频率分析的一些代表性研究成果,主要集中在以下两方面:一是基于还原/还现途径;二是基于非一致性极值系列直接进行水文频率分析途径。最后对进一步值得研究的问题进行了展望。

基于GIS的热带雨林地区滑坡敏感性分析——马来西亚金马伦高原个案研究

滑坡敏感性分析对灾害评价和预测具有重要的作用。以马来西亚金马伦高原为研究区,选择坡度、坡向、地表曲率、岩性、构造、土地覆盖、地貌类型、道路和排水系统等9个要素作为评价因子,探讨运用G IS和RS技术获取、管理滑坡灾害信息,和热带雨林地区湿热环境下滑坡灾害敏感性的分析方法。条件概率模型和逻辑回归模型分别应用于滑坡灾害敏感性分析与制图,通过比较滑坡敏感性的计算结果与历史滑坡信息,验证了两种方法的有效性,结果显示,条件概率模型和逻辑回归模型的预测精度分别为77.3%和83.6%,逻辑回归法具有较好的描述精度;滑坡敏感性分析中土地利用和土地覆盖、道路设施等因素具有较高权重,人类对雨林的垦殖和开发提升了该地区滑坡发生的敏感度。

基于状态条件概率的设备剩余寿命预测

针对机载设备剩余使用寿命预测中存在的不确定性因素,建立了基于状态条件概率分布的机载设备剩余寿命模型.首先,引入状态条件概率矢量对隐马尔科夫模型(HMM,Hidden Markov Model)进行不确定性改进,并推导了其计算形式.其次,给出了近似确定状态退化转移时间的方法,由此得到了以状态条件概率矢量为协变量的条件可靠度函数及剩余寿命模型.最后,以飞机发动机温控放大器为应用对象进行仿真计算.仿真结果表明该模型预测精度高,能够较大程度地降低不确定性因素的影响.

风电场输出功率的多时段联合概率密度预测

风电场输出功率波动性较强,难以精确预测,掌握其输出功率的分布规律对含有风电场的电力系统的运行决策具有重要意义。文中在分析风电场有功功率输出特性的基础上,提出了风电场输出功率多时段联合概率密度预测,利用风电场输出功率在时段间较强的相关性,估计其波动的幅度与速度特征,为系统运行提供更全面的决策信息。结合多元回归估计常条件相关—多元广义自回归条件异方差(CCC-MGARCH)模型与稀疏贝叶斯学习方法,给出了一种基于数值天气预报信息的风电场输出功率短期多时段联合概率密度预测方法。该方法依据CCC-MGARCH模型思想,将未来多个时段内风电场输出功率的联合概率密度预测问题分解为:风电场在各个时段内独立的输出功率概率密度预测子问题和时段间关联的输出功率预测误差相关系数矩阵估计子问题,利用稀疏贝叶斯学习方法在概率密度预测问题上的优势,形成预测效果好、计算效率高的风电场输出功率多时段联合概率密度预测方法。应用实例与分析说明了该方法的有效性。

任务情境和证据信息对青少年贝叶斯推理的影响研究

采用生活情境测查任务和经典测查任务探讨了任务情境对青少年贝叶斯推理的影响,以及生活情境测查任务中不同证据信息对青少年贝叶斯判断的影响作用。结果表明:(1)在生活情境测查任务中,从小学六年级到大学二年级,被试的贝叶斯推理能力稳步缓慢提升(相邻的两个被试组之间差异不显著,而不相邻的两个被试组之间差异更容易达到显著水平),发展没有出现加速期,而经典测查任务情境下没有表现出年龄差异;(2)贝叶斯推理作为条件概率的判断不仅受任务情境的影响,而且同一任务情境中不同证据信息也会影响贝叶斯判断。当证据信息与先验信念一致时,被试可以充分利用线索进行推理。

基于系统评价的块体可靠度指标分析及应用

利用条件概率评价及系统工程理论,将关键块体作为块体系统一个重要组成部分,分析了关键块体可靠度指标评价办法,探讨了关键块体与次生关键块可靠度内在关系,由此类推,提出块体系统评价方法,推导出块体系统可靠度计算公式,实现对块体系统的失稳概率的评价计算。并选择一工程实例,对评价方法予以验证。结果表明:传统关键块体可靠度评价办法忽略块体间内在影响,结果偏向风险;而文中方法综合考虑块体之间影响,并建立了系统评价模型,对块体系统评价更接近实际情况,对块体系统理论的研究有一定借鉴意义。