Establishing decision tree-based short-term default credit risk assessment models

ABSTRACT

Traditional credit risk assessment models do not consider the time factor; they only think of whether a customer will default, but not the when to default. The result cannot provide a manager to make the profit-maximum decision. Actually, even if a customer defaults, the financial institution still can gain profit in some conditions. Nowadays, most research applied the Cox proportional hazards model into their credit scoring models, predicting the time when a customer is most likely to default, to solve the credit risk assessment problem. However, in order to fully utilize the fully dynamic capability of the Cox proportional hazards model, time-varying macroeconomic variables are required which involve more advanced data collection. Since short-term default cases are the ones that bring a great loss for a financial institution, instead of predicting when a loan will default, a loan manager is more interested in identifying those applications which may default within a short period of time when approving loan applications. This paper proposes a decision tree-based short-term default credit risk assessment model to assess the credit risk. The goal is to use the decision tree to filter the short-term default to produce a highly accurate model that could distinguish default lending. This paper integrates bootstrap aggregating (Bagging) with a synthetic minority over-sampling technique (SMOTE) into the credit risk model to improve the decision tree stability and its performance on unbalanced data. Finally, a real case of small and medium enterprise loan data that has been drawn from a local financial institution located in Taiwan is presented to further illustrate the proposed approach. After comparing the result that was obtained from the proposed approach with the logistic regression and Cox proportional hazards models, it was found that the classifying recall rate and precision rate of the proposed model was obviously superior to the logistic regression and Cox proportional hazards models.     

川中丘陵区小麦品种绵阳26号的产量构成因素分析

运用相关、通径分析的方法分析了四川省小麦主推品种绵阳 2 6号的产量及其构成因素之间的关系。结果表明 ,绵阳 2 6号小麦品种千粒重的变异不大 ,穗粒数是产量构成的最重要因素 ,合理的有效穗是高产的前提。     

商业企业自有品牌战略研究

深入分析品牌的作用和自有品牌的特性,并就自有品牌的创造、服务保障、法律保护等提出若干对策。     

Additive and Multiplicative Models and Multistage Carcinogenesis Theory

In light of the Armitage-Doll multistage carcinogenesis theory, this paper examines the assumption that an additive relative risk relationship is indicative of two carcinogens that affect the same stage in the cancer process. We present formulas to compute excess cancer risks for a variety of patterns for limited exposure durations to two carcinogens that affect the first and penultimate stages; and using an index of synergy proposed by Thomas (1982), we find a number of these patterns to produce additive, or nearly additive, relative risk relationships. The consistent feature of these patterns is that the two exposure periods are of short duration and occur close together.     

《圆音正考》索解

"尖团音"原本是指满文的尖头字母和圆头字母,分别对应汉语的见组字和精组字.《圆音正考》为了区别业已合流的见精组细音字,对满汉对音规则作出调整,利用满文字头形状的不同将汉语的见组细音字和精组细音字分开,由此"尖团音"内涵变为见精组细音字.《圆音正考》"纠正"了前期满汉对音中见组细音字混乱的局面,即所谓"正考",这一作法适应了满汉对音的需要,因此影响深远取得尊位.     

基于语义生成与语义分割的机织物缺陷检测方法

针对织物疵点的语义分割任务中因数据集规模限制，而导致网络出现的严重过拟合问题，课题组提出了针对织物的语义生成网络。语义生成网络使用随机产生的语义标签生成对应的织物缺陷图像，相较于传统数据增强方法，语义生成可生成全新图像，更贴近实际缺陷分布，并且可通过判别器对生成图像进行筛选；课题组将语义生成的图像作为语义分割网络的输入，相应的随机语义标签作为目标，免去标注过程，扩充语义分割网络的训练样本，提升网络性能；对于语义分割网络，提出尺寸自适应Dice损失函数，解决样本不平衡问题，提升网络对小尺寸的检测能力。实验结果表明：尺寸自适应Dice损失函数使得模型精度提高11.1%，使用BEGAN扩充的数据集相较于传统方法扩充的数据集训练得到的模型精度提高7.4%。     

Quality Provision with Heterogeneous Consumer Reservation Utilities

This study examines a firm's quality and price decisions when consumers differ not only in their willingness‐to‐pay for quality but also in their reservation utility for the basic product. We find that while the firm offers lower‐quality products when consumers' valuations for quality deteriorate, the optimal quality may increase with a negative shift in consumers' reservation utilities. We also investigate the optimal price and quality of the products within a vertically differentiated product line when the number of products is exogenously given. The existing literature shows that when consumers differ only in their willingness‐to‐pay for quality, the firm sets the efficient quality for consumers with the highest valuation for quality, whereas the concern for cannibalization pushes down the quality of inferior products. We find that when consumers are heterogeneous in both their reservation utility and valuation for quality, the concern for cannibalization may distort the quality upwards, even for consumers with the highest willingness‐to‐pay for quality. In addition, a low‐quality product may enjoy a higher profit margin than a high‐quality product within the product line.     

赵树理文学的表征危机

用表征理论省察、考量赵树理文学,会发现赵树理文学也存在着表征危机,具体表现为其作品中的"次要人物"如二诸葛、三仙姑、小飞蛾的文学性描写远胜于其"主要人物"如小二黑、小芹、艾艾等,而这些人物又是他所着力否定的。其主观追求同实际效果之间发生了严重的背反。造成赵树理文学的表征危机的原因是:缺乏相应的谱系、话语范型支持,同时也与他对民间"小传统"中的诗意存在的回避、拒绝、否定有关。结果:一、造成对民间小传统的"暴力改写"和严重的妖魔化;二、使非文学的政治理性不断在文学土地上强势"殖民",导致文学越来越可能趋向被政治简单改写的危机。     

关于文学在魏晋南北朝的地位问题——兼析“文章与道德相抗衡”“文学高于一切”

魏晋南北朝时人已经清楚地认识到了文学本身具有不依附他者的价值,并对这一价值给以肯定。这标志着文学进入了自觉时代,标志着文学取得了独立的地位。然而独立后的文学并没有在东晋被谁真的认为过“驾乎道德之上”,更没有在南朝取得过“高于一切说”的地位。纵观历史,这不仅东晋南朝为然,而且其后的唐、宋、元、明、清等各时代也莫不如此。