基于贝叶斯方法的信用评级模型构建与违约概率估计

以贝叶斯方法为基础构建了信用评级和违约概率模型,指出金融机构利用已有评级信息提高债务人信用风险评估准确性的途径,并以单个债务人违约概率度量方法和Merton理论为基础,考虑异质性导致的宏观经济冲击对债务人的不同影响,度量资产组合违约风险。利用相关数据对贝叶斯模型应用给出例证,结果表明贝叶斯方法具有更为灵活的框架和较好的预测能力。     

潜在因素模型在商业银行信用风险分析中的应用

在评估商业银行整体信用风险时,债务人的信息一般不会传递到风险管理部门,导致在缺少违约数据时传统方法的分析十分复杂甚至难以进行.基于贝叶斯方法的潜在因素模型可以有效解决无法获得特定债务人信用质量的问题,并能够在宏观经济环境变动时准确评估违约风险强度变化,从而避免低估风险.利用MCMC模拟方法对商业银行数据的实证分析表明,潜在因素模型不仅推断方法及模拟途径简洁清晰,估计结果更加精确,而且在贝叶斯框架下具有较强的灵活性,适合在不同的数据约束条件下应用,便于国内风险分析人员采用.     

基于分层广义线性模型的非寿险费率厘定精算模型研究

非寿险业务中的损失数据结构日益复杂,呈现异质性与相关性并存的异象。分层广义线性模型能够突破传统费率厘定精算方法仅分析风险个体同一保单年损失数据的局限,可以提高复杂结构损失数据预测的准确性。基于分层广义线性模型等方法,研究具有多年损失数据的非寿险费率厘定问题,并以车险和工伤补偿保险的两组损失数据为例进行实证分析。研究结果表明,相对于GLM而言,考虑随机效应后GLMM的拟合优度大幅改善,GLMM与HGLM可以更有效地反映不同风险个体的差异,并有利于揭示风险个体在多个保险期内损失的异质性与相关性。     

大数据背景下贝叶斯模型平均的理论突破与应用前景

大数据统计分析过程中常面临模型比较和选择的不确定性问题。贝叶斯模型平均(BMA)方法可以通过先验和后验概率度量模型不确定性,并利用后验概率对模型的结果进行加权平均,最终得到更稳健的估计结果。在回顾贝叶斯模型平均发展历程的基础上,介绍贝叶斯模型平均的基本原理,综述其在一些难点问题上的理论进展,并介绍大数据背景下贝叶斯模型平均的应用前景。贝叶斯模型平均与复杂数据分析方法相结合,可能成为大数据研究的新思路。     

奈特不确定性下的贝叶斯学习行为模型

奈特不确定性和风险下的决策框架差异明显,传统风险度量模型在处理不确定性问题上表现的并不令人满意.处理奈特不确定性的主要统计决策方法可分为新古典方法、参数修正方法和贝叶斯方法.在贝叶斯框架下,可以基于学习行为构造一个融合外部环境信息和内生主观预期的模型,将期望效用最高的行为视为最优决策.贝叶斯学习行为模型在实践中能够有效的区分风险和不确定性,对于解决不确定性环境下传统风险度量模型容易低估风险和忽略极端风险的问题具有积极意义.     

基于MCMC稳态模拟的异质性经济增长模型研究

 针对同质性增长模型无法描述各个经济体经济增长存在的异质性现象,提出了一类基于MCMC稳态模拟的异质性经济增长模型,它可以用来描述经济增长的异质性以及政策变量的差异影响。由于模型参数的后验条件分布没有确定的分布形式,通过数据扩充得到参数的完全条件分布从而实现模型参数的贝叶斯估计。对改革开放以来我国各省市区经济增长收敛性进行分析发现大规模股份制改革前经济增长具有同质性而大规模股份制改革后经济增长具有异质性,且可以用新古典经济增长理论来解释各地区经济的发展状况。     

行政记录整合的贝叶斯分层记录链接模型及应用

记录链接的技术问题与统计理论密切相关,尤其是在建立记录链接分类规则时需要构建统计模型,识别关键变量以完成数据匹配。在贝叶斯框架下构建分层模型整合行政记录,通过多元回归可以实现匹配错误率的估计,而且一对一限制下的记录链接允许通过模块反映记录信息的来源变化,基于MCMC模拟的后验分布计算方便,有助于提高数据整合效率。     

基于Gibbs抽样的贝叶斯超高频金融数据协整关系研究

针对传统协整检验不能适用于具有随机性特征的超高频金融数据的问题,构建贝叶斯超高频金融数据协整模型,结合参数的后验条件分布设计Gibbs抽样方案,提出基于超高频金融数据的贝叶斯协整检验方法,并利用中国股市超高频金融数据进行实证分析。研究结果表明：贝叶斯方法把参数看作随机变量的思想适合超高频数据随机性的特点,贝叶斯超高频数据协整方法能够不断更新参数信息,避免了OLS估计的有偏性问题,可以得到更符合实际的结论。     

资产配置模型的选择:回报、风险抑或二者兼具

近年来以风险平价为代表的基于风险的配置模型广为流行。这些模型的一大特点是放弃回报信息。而以均值方差模型代表的基于回报的配置模型则认为回报很重要而且默认对回报的预测是准确的。这两种做法都有问题。考虑到回报的可预测性得到了大量经验研究的支持,那么对于基于风险的配置模型而言,完全放弃回报则意味着有关回报的有用信息得不到充分利用。对于基于回报的配置模型而言,不考虑参数估计误差而且对输入参数敏感的缺点也大大抵消了它们利用回报信息带来的好处。那么,回报是否重要以及应该如何使用回报成了资产配置研究所面临的一个重大问题。为此,本文提出以风险平价为配置基准,以贝叶斯VAR回报预测为主观观点的Black-Litterman(贝叶斯BL)模型回答这一命题。利用1952-2016年的美国股票和债券季度数据,本文将贝叶斯BL模型与现有配置模型进行比较研究。实证结果表明,相比基于回报的配置模型,贝叶斯BL模型降低了组合风险;相比基于风险的配置模型,贝叶斯BL模型增强了组合回报。这些特性来自于它既能利用回报可预测性带来的有用信息,又能够发挥基于风险的配置模型在控制风险方面的优势。因此该模型表现出增强回报和控制风险兼具的特点,是一条具有潜力的资产配置新方案。     

基于Dirichlet过程的非参数贝叶斯方法研究综述

基于Dirichlet过程的非参数贝叶斯方法将先验分布有效扩展到了非参数分布,并广泛应用于密度估计、分层线性模型、有序数据以及生存数据等领域的分析。由于理论和实际的需要,该方法不断发展。为此,文章综述了基于Dirichlet过程的非参数贝叶斯方法的构造、算法及其在不同理论框架下的最新进展,并展望了未来的几个发展方向。     

Describing the Dynamics of Attention to TV Commercials: A Hierarchical Bayes Analysis of the Time to Zap an Ad

This paper provides insights into the dynamics of attention to TV commercials via an analysis of the length of time that commercials are viewed before being 'zapped'. The model, which incorporates a flexible baseline hazard rate and captures unobserved heterogeneity across both consumers and commercials using a hierarchical Bayes approach, is estimated on two datasets in which commercial viewing is captured by a passive online device that continually monitors a household's TV viewing. Consistent with previous findings in psychology about the nature of attentional engagement in TV viewing, baseline hazard rates are found to be non-monotonic. In addition, the data show considerable ad-to-ad and household-to-household heterogeneity in zapping behavior. While one of the datasets contains some information on characteristics of the ads, these data do not reveal any firm links between the ad heterogeneity and the ad characteristics. A number of methodological and computational issues arise in the hierarchical Bayes analysis.     

Analysis of Between–Household Heterogeneity in Disease Transmission from Data on Outbreak Sizes

The paper proposes a method of analysis for data on within–household disease transmission, when only outbreak sizes are available. The method assumes between–household heterogeneity of the transmission probabilities. A random effects model in a hierarchical setting is fitted using MCMC and data augmentation techniques. The procedure is illustrated on a measles dataset.     

A Bayesian approach of joint models for clustered zero-inflated count data with skewness and measurement errors

Count data with excess zeros are widely encountered in the fields of biomedical, medical, public health and social survey, etc. Zero-inflated Poisson (ZIP) regression models with mixed effects are useful tools for analyzing such data, in which covariates are usually incorporated in the model to explain inter-subject variation and normal distribution is assumed for both random effects and random errors. However, in many practical applications, such assumptions may be violated as the data often exhibit skewness and some covariates may be measured with measurement errors. In this paper, we deal with these issues simultaneously by developing a Bayesian joint hierarchical modeling approach. Specifically, by treating intercepts and slopes in logistic and Poisson regression as random, a flexible two-level ZIP regression model is proposed, where a covariate process with measurement errors is established and a skew-t-distribution is considered for both random errors and random effects. Under the Bayesian framework, model selection is carried out using deviance information criterion (DIC) and a goodness-of-fit statistics is also developed for assessing the plausibility of the posited model. The main advantage of our method is that it allows for more robustness and correctness for investigating heterogeneity from different levels, while accommodating the skewness and measurement errors simultaneously. An application to Shanghai Youth Fitness Survey is used as an illustrate example. Through this real example, it is showed that our approach is of interest and usefulness for applications.     

On the clustering term in ecological analysis: how do different prior specifications affect results?

We study how different prior assumptions on the spatially structured heterogeneity term of the convolution hierarchical Bayesian model for spatial disease data could affect the results of an ecological analysis when response and exposure exhibit a strong spatial pattern. We show that in this case the estimate of the regression parameter could be strongly biased, both by analyzing the association between lung cancer mortality and education level on a real dataset and by a simulation experiment. The analysis is based on a hierarchical Bayesian model with a time dependent covariate in which we allow for a latency period between exposure and mortality, with time and space random terms and misaligned exposure-disease data.     

我国宏观金融风险测度研究

文章根据统计指标体系的构建原则,建立了由宏观经济环境、银行呆坏账、泡沫、国债和外资冲击在内的宏观金融风险测度指标体系,利用映射法原则将原始指标区间化。然后根据层次分析法确定风险权重,进而构造了我国宏观金融风险测度的理论模型。     

A local join counts methodology for spatial clustering in disease from relative risk models

ABSTRACT

This paper considers adaptation of hierarchical models for small area disease counts to detect disease clustering. A high risk area may be an outlier (in local terms) if surrounded by low risk areas, whereas a high risk cluster requires that both the focus area and surrounding areas demonstrate common elevated risk. A local join count method is suggested to detect local clustering of high disease risk in a single health outcome, and extends to assessing bivariate spatial clustering in relative risk. Applications include assessing spatial heterogeneity in effects of area predictors according to local clustering configuration, and gauging sensitivity of bivariate clustering to random effect assumptions.     

Hierarchical overdispersed Poisson model with macrolevel autocorrelation

We review Bayesian analysis of hierarchical non-standard Poisson regression models with an emphasis on microlevel heterogeneity and macrolevel autocorrelation. For the former case, we confirm that negative binomial regression usually accounts for microlevel heterogeneity (overdispersion) satisfactorily; for the latter case, we apply the simple first-order Markov transition model to conveniently capture the macrolevel autocorrelation which often arises from temporal and/or spatial count data, rather than attaching complex random effects directly to the regression parameters. Specifically, we extend the hierarchical (multilevel) Poisson model into negative binomial models with macrolevel autocorrelation using restricted gamma mixture with unit mean and Markov transition covariate created from preceding residuals. We prove a mild sufficient condition for posterior propriety under flat prior for the interesting fixed effects. Our methodology is implemented by analyzing the Baltic sea peracarids diurnal activity data published in the marine biology and ecology literature.     

Spatial measurement error in infectious disease models

Individual-level models (ILMs) for infectious disease can be used to model disease spread between individuals while taking into account important covariates. One important covariate in determining the risk of infection transfer can be spatial location. At the same time, measurement error is a concern in many areas of statistical analysis, and infectious disease modelling is no exception. In this paper, we are concerned with the issue of measurement error in the recorded location of individuals when using a simple spatial ILM to model the spread of disease within a population. An ILM that incorporates spatial location random effects is introduced within a hierarchical Bayesian framework. This model is tested upon both simulated data and data from the UK 2001 foot-and-mouth disease epidemic. The ability of the model to successfully identify both the spatial infection kernel and the basic reproduction number (R ₀) of the disease is tested.