Conditional autoregressive negative binomial model for analysis of crash count using Bayesian methods

In order to improve crash occurrence models to account for the influence of various contributing factors, a conditional autoregressive negative binomial （CAR-NB） model is employed to allow for overdispersion （tackled by the NB component）, unobserved heterogeneity and spatial autocorrelation （captured by the CAR process）, using Markov chain Monte Carlo methods and the Gibbs sampler. Statistical tests suggest that the CAR-NB model is preferred over the CAR-Poisson, NB, zero-inflated Poisson, zero-inflated NB models, due to its lower prediction errors and more robust parameter inference. The study results show that crash frequency and fatalities are positively associated with the number of lanes, curve length, annual average daily traffic （AADT） per lane, as well as rainfall. Speed limit and the distances to the nearest hospitals have negative associations with segment-based crash counts but positive associations with fatality counts, presumably as a result of worsened collision impacts at higher speed and time loss during transporting crash victims.

A Slice Analysis-Based Bayesian Inference Dynamic Power Model for CMOS Combinational Circuits

To improve the accuracy and speed in cycle-accurate power estimation, this paper uses multiple dimensional coefficients to build a Bayesian inference dynamic power model. By analyzing the power distribution and internal node state, we find the deficiency of only using port information. Then, we define the gate level number computing method and the concept of slice, and propose using slice analysis to distill switching density as coefficients in a special circuit stage and participate in Bayesian inference with port information. Experiments show that this method can reduce the power-per-cycle estimation error by 21.9% and the root mean square error by 25.0% compared with the original model, and maintain a 700 ＋ speedup compared with the existing gate-level power analysis technique.

Local and regional flood frequency analysis based on hierarchical Bayesian model in Dongting Lake Basin,China

This study developed a hierarchical Bayesian(HB)model for local and regional flood frequency analysis in the Dongting Lake Basin,in China.The annual maximum daily flows from 15 streamflow-gauged sites in the study area were analyzed with the HB model.The generalized extreme value(GEV)distribution was selected as the extreme flood distribution,and the GEV distribution location and scale parameters were spatially modeled through a regression approach with the drainage area as a covariate.The Markov chain Monte Carlo(MCMC)method with Gibbs sampling was employed to calculate the posterior distribution in the HB model.The results showed that the proposed HB model provided satisfactory Bayesian credible intervals for flood quantiles,while the traditional delta method could not provide reliable uncertainty estimations for large flood quantiles,due to the fact that the lower confidence bounds tended to decrease as the return periods increased.Furthermore,the HB model for regional analysis allowed for a reduction in the value of some restrictive assumptions in the traditional index flood method,such as the homogeneity region assumption and the scale invariance assumption.The HB model can also provide an uncertainty band of flood quantile prediction at a poorly gauged or ungauged site,but the index flood method with L-moments does not demonstrate this uncertainty directly.Therefore,the HB model is an effective method of implementing the flexible local and regional frequency analysis scheme,and of quantifying the associated predictive uncertainty.

Bayesian analysis of minimal model under the insulin-modified IVGTT

A Bayesian analysis of the minimal model was proposed where both glucose and insulin were analyzed simultaneously under the insulin-modified intravenous glucose tolerance test (IVGTT). The resulting model was implemented with a nonlinear mixed-effects modeling setup using ordinary differential equations (ODEs), which leads to precise estimation of population parameters by separating the inter- and intra-individual variability. The results indicated that the Bayesian method applied to the glucose-insulin minimal model provided a satisfactory solution with accurate parameter estimates which were numerically stable since the Bayesian method did not require approximation by linearization.

Bayesian Network and Factor Analysis for Modeling Pine Wilt Disease Prevalence

A Bayesian network (BN) model was developed to predict susceptibility to PWD(Pine Wilt Disease). The distribution of PWD was identified using QuickBird and unmanned aerial vehicle (UAV) images taken at different times. Seven factors that influence the distribution of PWD were extracted from the QuickBird images and were used as the independent variables. The results showed that the BN model predicted PWD with high accuracy. In a sensitivity analysis, elevation (EL), the normal differential vegetation index (NDVI), the distance to settlements (DS) and the distance to roads (DR) were strongly associated with PWD prevalence, and slope (SL) exhibited the weakest association with PWD prevalence. The study showed that BN is an effective tool for modeling PWD prevalence and quantifying the impact of various factors.

Bayesian meta-analysis of regional biomass factors for Quercus mongolica forests in South Korea

Indirect approaches to estimation of biomass factors are often applied to measure carbon flux in the forestry sector. An assumption underlying a country-level carbon stock estimate is the representativeness of these factors. Although intensive studies have been conducted to quantify biomass factors, each study typically covers a limited geographic area. The goal of this study was to employ a meta-analysis approach to develop regional bio- mass factors for Quercus mongolica forests in South Korea. The biomass factors of interest were biomass conversion and expansion factor （BCEF）, biomass expansion factor （BEF） and root-to-shoot ratio （RSR）. Our objectives were to select probability density functions （PDFs） that best fitted the three biomass factors and to quantify their means and uncertainties. A total of 12 scientific publications were selected as data sources based on a set of criteria. Fromthese publications we chose 52 study sites spread out across South Korea. The statistical model for the meta- analysis was a multilevel model with publication （data source） as the nesting factor specified under the Bayesian framework. Gamma, Log-normal and Weibull PDFs were evaluated. The Log-normal PDF yielded the best quanti- tative and qualitative fit for the three biomass factors. However, a poor fit of the PDF to the long right tail of observed BEF and RSR distributions was apparent. The median posterior estimates for means and 95 % credible intervals for BCEF, BEF and RSR across all 12 publica- tions were 1.016 （0.800-1.299）, 1.414 （1.304-1.560） and 0.260 （0.200-0.335）, respectively. The Log-normal PDF proved useful for estimating carbon stock of Q. mongolica forests on a regional scale and for uncertainty analysis based on Monte Carlo simulation.

Discrimination for minimal hepatic encephalopathy based on Bayesian modeling of default mode network

In order to classify the minimal hepatic encephalopathy （MHE） patients from healthy controls, the independent component analysis （ICA） is used to generate the default mode network （DMN） from resting-state functional magnetic resonance imaging （fMRI）. Then a Bayesian voxel- wised method, graphical-model-based multivariate analysis （GAMMA）, is used to explore the associations between abnormal functional integration within DMN and clinical variable. Without any prior knowledge, five machine learning methods, namely, support vector machines （SVMs）, classification and regression trees （ CART ）, logistic regression, the Bayesian network, and C4.5, are applied to the classification. The functional integration patterns were alternative within DMN, which have the power to predict MHE with an accuracy of 98%. The GAMMA method generating functional integration patterns within DMN can become a simple, objective, and common imaging biomarker for detecting MIIE and can serve as a supplement to the existing diagnostic methods.

Bayesian Joint Modelling of Survival Time and Longitudinal CD4 Cell Counts Using Accelerated Failure Time and Generalized Error Distributions

Survival of HIV/AIDS patients is crucially dependent on comprehensive and targeted medical interventions such as supply of antiretroviral therapy and monitoring disease progression with CD4 T-cell counts. Statistical modelling approaches are helpful towards this goal. This study aims at developing Bayesian joint models with assumed generalized error distribution (GED) for the longitudinal CD4 data and two accelerated failure time distributions, Lognormal and loglogistic, for the survival time of HIV/AIDS patients. Data are obtained from patients under antiretroviral therapy follow-up at Shashemene referral hospital during January 2006-January 2012 and at Bale Robe general hospital during January 2008-March 2015. The Bayesian joint models are defined through latent variables and association parameters and with specified non-informative prior distributions for the model parameters. Simulations are conducted using Gibbs sampler algorithm implemented in the WinBUGS software. The results of the analyses of the two different data sets show that distributions of measurement errors of the longitudinal CD4 variable follow the generalized error distribution with fatter tails than the normal distribution. The Bayesian joint GED loglogistic models fit better to the data sets compared to the lognormal cases. Findings reveal that patients’ health can be improved over time. Compared to the males, female patients gain more CD4 counts. Survival time of a patient is negatively affected by TB infection. Moreover, increase in number of opportunistic infection implies decline of CD4 counts. Patients’ age negatively affects the disease marker with no effects on survival time. Improving weight may improve survival time of patients. Bayesian joint models with GED and AFT distributions are found to be useful in modelling the longitudinal and survival processes. Thus we recommend the generalized error distributions for measurement errors of the longitudinal data under the Bayesian joint modelling. Further studies may investigate the models with various types of shared random effects and more covariates with predictions.

Minimum Description Length Methods in Bayesian Model Selection: Some Applications

Computations involved in Bayesian approach to practical model selection problems are usually very difficult. Computational simplifications are sometimes possible, but are not generally applicable. There is a large literature available on a methodology based on information theory called Minimum Description Length (MDL). It is described here how many of these techniques are either directly Bayesian in nature, or are very good objective approximations to Bayesian solutions. First, connections between the Bayesian approach and MDL are theoretically explored;thereafter a few illustrations are provided to describe how MDL can give useful computational simplifications.

Analysis of State Homicide Rates Using Statistical Ranking and Selection Procedures

Nonparametric and parametric subset selection procedures are used in the analysis of state homicide rates (SHRs), for the year 2005 and years 2014-2020, to identify subsets of states that contain the “best” (lowest SHR) and “worst” (highest SHR) rates with a prescribed probability. A new Bayesian model is developed and applied to the SHR data and the results are contrasted with those obtained with the subset selection procedures. All analyses are applied within the context of a two-way block design.

Estimation of the Piecewise Exponential Model by Bayesian P-Splines via Gibbs Sampling: Robustness and Reliability of Posterior Estimates

In the investigation of disease dynamics, the effect of covariates on the hazard function is a major topic. Some recent smoothed estimation methods have been proposed, both frequentist and Bayesian, based on the relationship between penalized splines and mixed models theory. These approaches are also motivated by the possibility of using automatic procedures for determining the optimal amount of smoothing. However, estimation algorithms involve an analytically intractable hazard function, and thus require ad-hoc software routines. We propose a more user-friendly alternative, consisting in regularized estimation of piecewise exponential models by Bayesian P-splines. A further facilitation is that widespread Bayesian software, such as WinBUGS, can be used. The aim is assessing the robustness of this approach with respect to different prior functions and penalties. A large dataset from breast cancer patients, where results from validated clinical studies are available, is used as a benchmark to evaluate the reliability of the estimates. A second dataset from a small case series of sarcoma patients is used for evaluating the performances of the PE model as a tool for exploratory analysis. Concerning breast cancer data, the estimates are robust with respect to priors and penalties, and consistent with clinical knowledge. Concerning soft tissue sarcoma data, the estimates of the hazard function are sensitive with respect to the prior for the smoothing parameter, whereas the estimates of regression coefficients are robust. In conclusion, Gibbs sampling results an efficient computational strategy. The issue of the sensitivity with respect to the priors concerns only the estimates of the hazard function, and seems more likely to occur when non-large case series are investigated, calling for tailored solutions.

Quasi-Bayesian software reliability model with small samples

In traditional Bayesian software reliability models, it was assume that all probabilities are precise. In practical applications the parameters of the probability distributions are often under uncertainty due to strong dependence on subjective information of experts＇ judgments on sparse statistical data. In this paper, a quasi-Bayesian software reliability model using interval-valued probabilities to clearly quantify experts＇ prior beliefs on possible intervals of the parameters of the probability distributions is presented. The model integrates experts＇ judgments with statistical data to obtain more convincible assessments of software reliability with small samples. For some actual data sets, the presented model yields better predictions than the Jelinski-Moranda （JM） model using maximum likelihood （ML）.

基于贝叶斯理论的TOPMODEL参数不确定性分析

应用贝叶斯理论探讨了流域水文模型参数及预报不确定性问题.通过Monte-Carlo途径确定TOPMODEL模型的敏感参数,采用MCMC抽样技术估计敏感参数的后验概率密度分布,并根据参数的抽样系列构造水文模型预报值的经验分布,据此对模型参数的不确定性及其对水文预报结果的影响进行评价.以浙江密赛流域为例进行了应用研究,提供了模型参数及预报结果不确定性的定量分析结果.

研制阶段系统可靠性增长的Bayesian评估与预测

基于新Dirichlet先验分布,建立一种适合小子样复杂系统异总体可靠性增长分析的Bayesian模型。充分利用先验信息和阶段试验信息,结合产品研制的试验数据,利用最优化方法研究新的Dirichlet先验分布容易定量和衡量先验参数确定的方法,解决了超参数物理意义不明确难以确定问题。通过变量替换的Gibbs抽样简化了后验推断,合理估算出当前阶段和后续试验阶段产品可靠性的Bayesian点估计和置信下限;结合试验数据,利用该模型实现了未来阶段可靠性的预测,扩展了模型应用范围。实例表明该模型参数含义清晰明确,简单易行,利于工程应用。

Bayesian两变量层次模型及其在诊断试验系统评价中的应用

目的探讨Bayesian两变量层次模型的构建及其在诊断试验系统评价中的应用。方法将Bayesian两变量层次模型应用于传统Pap细胞学涂片诊断子宫颈癌准确性评价的历史Meta分析资料,估计相关的效应指标敏感度和特异度及筛查研究比随访研究的相对可信度。结果与经典综合受试者工作特征曲线方法相比,Bayesian两变量层次模型估计得到三个层次的效应指标,其中综合敏感度和特异度均数及95%可信区间分别为0.64(0.56,0.72)和0.74(0.67,0.80),预测敏感度和特异度均数及95%可信区间分别为0.61(0.12,0.96)和0.69(0.21,0.97),筛查研究比随访研究的相对可信度估计为1.3(0.59,2.48)。结论采用Bayesian两变量层次模型进行诊断试验Meta分析,更加灵活、有效,易于实现和解释,值得推广应用。

基于Bayesian层次时空模型的我国老龄化分析与预测

本文首次利用Bayesian层次时空模型,以1995—2014年全国省级人口统计数据为基础,分析了近20年来我国老龄化在空间和时间上的变化规律。研究发现:1我国高老龄化地区分布已形成X型地理空间分布结构,东部地区为主,西部地区为辅,总体老龄化率呈上升趋势;2四川、重庆、辽宁、安徽、湖北和湖南等6个地区不仅是老龄化热点区域,而且老龄化增速也快于全国平均水平,特别是四川和重庆,老龄化程度和增速都是全国最高;3中西部地区老龄化程度虽然低于全国平均水平,但增加速度却高于全国平均水平;4北京、天津、上海、江苏、浙江和广东等6个高老龄化地区的老龄化率趋于平稳或增速放缓;5预测"全面二孩"政策情境下我国2030年老龄化率为13.19%(11.10%,20.94%)。

改进Bayesian后验比的异常风速值检测方法

风电场运行数据中含有异常风速值,为了优化风电数据的质量,提出了组合预测与Bayesian后验比的异常值检测方法。为了降低预测误差,先对风速序列建立Adaboost-BP网络和EMD-LV-SVM的组合预测模型,利用预测值与测量值的偏差得到含有粗大误差的残差序列;为了提高检测方法的可靠性,采用Bayesian后验比的检验方法识别残差序列中粗大误差,从而确定异常风速值的位置,并利用ARIMA方法修正异常风速值。RBF预测结果表明,所提方法能准确识别异常值,从而提高了风电场短期风速预测精度。

Mathematical Modeling of Carcinogenesis Based on Chromosome Aberration Data

Objective： The progression of human cancer is characterized by the accumulation of genetic instability. An increasing number of experimental genetic molecular techniques have been used to detect chromosome aberrations. Previous studies on chromosome abnormalities often focused on identifying the frequent loci of chromosome alterations, but rarely addressed the issue of interrelationship of chromosomal abnormalities. In the last few years, several mathematical models have been employed to construct models of carcinogenesis, in an attempt to identify the time order and cause-and-effect relationship of chromosome aberrations. The principles and applications of these models are reviewed and compared in this paper. Mathematical modeling of carcinogenesis can contribute to our understanding of the molecular genetics of tumor development, and identification of cancer related genes, thus leading to improved clinical practice of cancer.

Objective Bayesian analysis for the accelerated degradation model usingWiener process with measurement errors

The Wiener process as a degradation model plays an important role in the degradation analysis.In this paper, we propose an objective Bayesian analysis for an acceleration degradation Wienermodel which is subjected to measurement errors. The Jeffreys prior and reference priors underdifferent group orderings are first derived, the propriety of the posteriors is then validated. It isshown that two of the reference priors can yield proper posteriors while the others cannot. A simulation study is carried out to investigate the frequentist performance of the approach comparedto the maximum likelihood method. Finally, the approach is applied to analyse a real data.

利用Bayesian方法进行牙鲆体质量和主要形态性状的QTL定位及互作分析

为了定位牙鲆Paralichthys olivaceus体质量和形态性状的QTL,利用雌核发育方法获得双单倍体牙鲆群体的157个个体,同时构建了包含481个SSR标记和141个EST-SSR标记的遗传连锁图谱,并采用Bayesian模型选择方法对研究性状进行主效应QTL定位和上位性互作分析。结果表明：共检测到38个主效应QTL和50对上位互作效应,其中,大部分主效应QTL位于5、6、9、22号染色体,上位互作效应所解释的表型方差为2.3%~23.1%;8个主效应QTL和7对互作QTL表现出多效性,均分别对两个以上的性状起作用。本研究结果将有助于提高牙鲆育种中对有益经济性状的选择效率。