Bayesian and Frequentist Inference for Ecological Inference: The R×C Case

In this paper we propose Bayesian and frequentist approaches to ecological inference, based on R × C contingency tables, including a covariate. The proposed Bayesian model extends the binomial-beta hierarchical model developed by K ing , R osen and T anner (1999) from the 2×2 case to the R × C case. As in the 2×2 case, the inferential procedure employs Markov chain Monte Carlo (MCMC) methods. As such, the resulting MCMC analysis is rich but computationally intensive. The frequentist approach, based on first moments rather than on the entire likelihood, provides quick inference via nonlinear least-squares, while retaining good frequentist properties. The two approaches are illustrated with simulated data, as well as with real data on voting patterns in Weimar Germany. In the final section of the paper we provide an overview of a range of alternative inferential approaches which trade-off computational intensity for statistical efficiency.     

The Macroeconomic Consequences of Mixing Sunnis and Shias: A Bayesian Errors-in-Variables Approach

How damaging is the Sunni/Shia split to the economies of Islamic countries? Is it better to be one or the other? Or is it better to have an even balance between the two? Answering these questions is complicated by the fact that the data are often missing or imprecisely measured. We employ the technique of Bayesian data augmentation to circumvent these two problems, and find that properly controlling for these features in the data leads to drastically different conclusions than what is found using ordinary least squares. We find that there is nothing in the differential nature of Sunni or Shia Islam to make one more economically prosperous than the other. Nor do we find any support for the popular hypothesis that Sunnis and Shias cannot mix.     

Stock selection using a multi-factor model - empirical evidence from the French stock market

Using a Barra-type factor model, we have attempted to determine whether it is possible to beat the benchmark by taking advantage of anomalies established in the financial empirical literature. More specifically we have built an equity premium model based on three sets of factors (accounting variables, stock market characteristics and sector indicators) using a Bayesian method corrected for heteroscedasticity to estimate risk premiums, a technique that takes agents' learning into account. The results are encouraging: first, the factors that carried most weight on the equity premiums corroborated the results of empirical studies described in the financial literature, secondly, the portfolios constructed from our methodology and simulated outside our sample, returned higher performance than the benchmark and rewarded the supplement of volatility.     

Interactions between real economic and financial sides of the US economy in a regime-switching environment

This objective of this study is to examine the linkages between real (economic) and financial variables in the United States in a regime-switching environment that accounts explicitly for high volatility in the stock market and high stress in financial markets. Since the linearity test shows that the linear model should be rejected, we employ the Markov-switching VECM to examine the same objective using the Bayesian Markov-chain Monte Carlo method. The regime-dependent impulse response function (RDIRF) highlights the increasing importance of the financial sector of the economy during stress periods. The responses and their fluctuations are significantly greater in the high-volatility regime than in the low-volatility regime.     

Estimating DSGE models across time and frequency

Using a wavelet-based decomposition, this paper exploits the information in the data usually employed in the estimation of DSGE models. A simple New Keynesian model featuring price and wage rigidities is estimated for the United States across different frequencies. The estimations indicate that most structural parameters exhibit a frequency-dependent behavior. The impulse response functions also indicate frequency-dependent responses of output to the exogenous shocks. For lower frequencies, there are more persistent effects, especially for preference and technology shocks.     

供应链风险评估:一个贝叶斯——量子决策的推理网络研究范式

为了分析供应链运作过程中存在风险的性质,采用贝叶斯网的方式,构建了供应链风险因素、风险事件之间的系统模型,并在量子概率的基础上,分析各种风险发生的规律,并分析了供应链风险事件之间存在的共同发生的可能性。结论认为:供应链风险是一个具有多源、多维、交叉等多种属性的复杂系统,其中,道德风险与牛鞭效应最容易发生,双重边际效应、资金链断裂以及牛鞭效应的影响最大;若干种风险之间按照一个较高的概率发生着共振现象。     

A forecast simulation analysis of the next‐generation DVD market based on consumer preference data

Just as standards wars over formats had characterized VCRs upon introduction to the market, the next‐generation DVD standards war between Blu‐Ray and HD‐DVD lasted 6 years before Blu‐Ray won the contest. Beginning with stated preference data drawn from a structured conjoint survey conducted before Blu‐Ray became the de facto standard of the next‐generation DVD format, we estimate consumer preferences on digital video players. A Bayesian mixed‐logit model is used and market share simulations are conducted under various scenarios based on the estimated parameters from Bayesian mixed‐logit model to surmise the future South Korean digital video‐player market. Results indicate that consumers feel that network size and title availability are more important than hardware‐related facets of the product, such as definition and storage capacity. The level of title availability and price of the Blu‐Ray player for Blu‐Ray's dominance over DVD will dramatically vary by the penetration rate of DVD players.     

数控刀具预调装置的项目教学法探索

项目教学法是工科类高校先进的教学手段,本文通过对数控刀具预调装置的课题阐述了对项目教学法的认识,以及如何在现代制造技术教学中应用项目教学法完成数控刀具预调装置的课题,从而不断完善现代制造技术教学体系。     

On the welfare costs of misspecified monetary policy objectives

This paper quantifies the effects on welfare of misspecified monetary policy objectives in a stylized DSGE model. We show that using inappropriate objectives generates relatively large welfare costs. When expressed in terms of ‘consumption equivalent’ units, these costs correspond to permanent decreases in steady-state consumption of up to two percent. The latter are generated by both the inappropriate choice of weights and the omission of variables. In particular, it is costly to assume an interest-rate smoothing incentive for central bankers when it is not socially optimal to do so. Finally, a parameter uncertainty decomposition indicates that uncertainty about the properties of markup shocks gives rise to the largest welfare costs.     

基于高级法的两维评级贷款审批标准

新资本协议框架下,公司暴露的风险加权资产依赖于客户的违约概率和债项的违约损失率。随着内部评级高级法的实施,商业银行将逐步建立和完善客户评级与债项评级体系,用于估算违约概率和违约损失率。本文深入分析了新资本协议的资本要求计算公式,提出以资本要求低于8%为导向、以客户评级和债项评级为基础的贷款两维准入标准。目的是帮助商业银行有效管理资本充足率,以及解决中小企业融资难等问题。