Ranking Crop Yield Models: A Comment

This comment discusses key specification issues that may have affected the performance and, therefore, the ranking of parametric models that were compared in a recent AJAE article. A procedure to obtain the most flexible parametric model specification possible, given the particular probability distribution function on which the model is based is presented. These specifications also allow for standardized and, therefore, more valid comparisons across parametric models that are based on different probability distributions. Finally, the comment cautions against generalization of the rankings in that AJAE article and recommends that these more flexible specifications be adopted in future comparisons and applications.     

计量模型中的加总偏误与内生性：一种数值模拟方法

理论文献已经证明了细分数据模型的内生性和加总过程产生的内生性是加总偏误的根本原因。但是由于内生性涉及误差项与回归量之间的相关性问题,试图通过实证方法审视这两类内生性对加总偏误的影响变得比较困难,而数值模拟却是一个较为理想的方法。在数值模拟中,我们通过控制随机变量的分布形式以及随机变量之间的相关程度,进而对两类内生性因素产生的加总偏误进行全面细致的考察。本文的研究将为加总偏误的内生性解释提供有力证据。     

Comparison of Hypothesis Testing and Bayesian Model Selection

The main goal of both Bayesian model selection and classical hypotheses testing is to make inferences with respect to the state of affairs in a population of interest. The main differences between both approaches are the explicit use of prior information by Bayesians, and the explicit use of null distributions by the classicists. Formalization of prior information in prior distributions is often difficult. In this paper two practical approaches (encompassing priors and training data) to specify prior distributions will be presented. The computation of null distributions is relatively easy. However, as will be illustrated, a straightforward interpretation of the resulting p-values is not always easy. Bayesian model selection can be used to compute posterior probabilities for each of a number of competing models. This provides an alternative for the currently prevalent testing of hypotheses using p-values. Both approaches will be compared and illustrated using case studies. Each case study fits in the framework of the normal linear model, that is, analysis of variance and multiple regression.     

Forecasting turning points in tourism growth

Tourism demand exhibits growth cycles, and it is important to forecast turning points in these growth cycles to minimise risks to destination management. This study estimates logistic models of Hong Kong tourism demand, which are then used to generate both short- and long-term forecasts of tourism growth. The performance of the models is evaluated using the quadratic probability score and hit rates. The results show that the ways in which this information is used are crucial to the models’ predictive power. Further, we investigate whether combining probability forecasts can improve predictive accuracy, and find that combination approaches, especially nonlinear combination approaches, are sensitive to the quality of forecasts in the pool. In addition, model screening can improve forecasting performance.     

证券投资风险预测的E-Bayes估计

本文给出了E-Bayes方法,以上海证券个股五粮液52个连续交易日的收盘价格为例,建立数学模型进行分析和预测,预测结果与市场实际值相当吻合。与灰色系统理论中的GM（1,1）预测模型相比,本文提出的方法预测的精度更高,计算量小。不仅适用于经济系统的分析与预测,也适用于其它系统的分析与预测。     

基于Eclipse的Scal模块实现毛管力和相渗数据合并的方法

在应用研究Eclipse软件的基础上,提出了利用Eclipse软件提供的Scal模块按照油藏模拟器要求把毛管力和相渗数据合并成一个新的以相饱和度相联系的饱和度函数表的新方法。实践表明,该方法简便实用。     

Forecasting and decomposition of portfolio credit risk using macroeconomic and frailty factors

This paper presents a dynamic portfolio credit model following the regulatory framework, using macroeconomic and latent risk factors to predict the aggregate loan portfolio loss in a banking system. The latent risk factors have three levels: global across the entire banking system, parent-sectoral for the intermediate loan sectors and sector-specific for the individual loan sectors. The aggregate credit loss distribution of the banking system over a risk horizon is generated by Monte Carlo simulation, and a quantile estimator is used to produce the aggregate risk measure and economic capital. The risk contributions of the individual sectors and risk factors are measured by combining the Hoeffding decomposition with the Euler capital allocation rule. For the U.S. banking system, we find that the real GDP growth rate, the global and sector-wide frailty risk factors and their spillovers significantly affect loan defaults, and the impacts of the frailty factors are not only economy-wide but also sector-specific. We also find that the frailty risk factors make more significant risk contributions to the aggregate portfolio risk than the macroeconomic factors, while the macroeconomic factors help to improve the accuracy and efficiency of the credit risk forecasts.     

PORTFOLIO CHOICE VIA QUANTILES

A portfolio choice model in continuous time is formulated for both complete and incomplete markets, where the quantile function of the terminal cash flow, instead of the cash flow itself, is taken as the decision variable. This formulation covers a wide body of existing and new models with law‐invariant preference measures, including expected utility maximization, mean–variance, goal reaching, Yaari's dual model, Lopes' SP/A model, behavioral model under prospect theory, as well as those explicitly involving VaR and CVaR in objectives and/or constraints. A solution scheme to this quantile model is proposed, and then demonstrated by solving analytically the goal‐reaching model and Yaari's dual model. A general property derived for the quantile model is that the optimal terminal payment is anticomonotonic with the pricing kernel (or with the minimal pricing kernel in the case of an incomplete market if the investment opportunity set is deterministic). As a consequence, the mutual fund theorem still holds in a market where rational and irrational agents co‐exist.     

客观现象的概率统计解释举例

本文通过应用概率统计方法解释人类行为与认识活动实例的分析,说明了基于概率统计方法如何描述客观现象和理解人类的认识行为,进而得出概率统计方法是基于归纳的演绎推理,其本质是认识论中的归纳推理。     

随机利率下的复合Poisson-Geometric风险模型及破产概率

本文主要研究的是在随机利率下保费收入为复合Poisson-Geometric过程的风险模型,在随机利率为levy过程的情况下,得到了破产概率满足的积分方程,以及得到最终破产概率的上下界所满足的积分不等式,以此作为保险公司经营的预警信号更具有现实意义。