基于粗糙集理论的决策树方法在贷款客户信用评估中的应用

决策树是数据挖掘中常用的分类技术，其生成的规则便于决策者理解和应用。然而面对较多的属性且含有冗余和噪声属性的记录集生成的决策树时，无法删除冗余属性，造成运算过程复杂。本文旨在通过应用粗糙集理论，将其与决策树方法进行结合，对属性进行约简，降低运算复杂度，并生成相对简化的规则形式，并将其应用到银行个人贷款客户信用评估之中。     

Relative Guarantees

Many real-world financial contracts have some sort of minimum rate of return guarantee included. One class of these guarantees is so-called relative guarantees, i.e., guarantees where the minimum guaranteed rate of return is given as a function of the stochastic return on a reference portfolio. These guarantees are the topic of this paper. We analyse a wide range of different functional specifications for the minimum guaranteed rate of return, hereunder both so-called maturity and multi-period guarantees. Several closed form solutions are presented.     

Cost monotonicity, consistency and minimum cost spanning tree games

We propose a new cost allocation rule for minimum cost spanning tree games. The new rule is a core selection and also satisfies cost monotonicity. We also give characterisation theorems for the new rule as well as the much-studied Bird allocation. We show that the principal difference between these two rules is in terms of their consistency properties.     

OPTIONS AND EFFICIENCY IN MULTIDATE SECURITY MARKETS

This paper extends the work of Ross (1976; Q. J. Econ. (90)1, 75–89) to multidate security markets. First, we show that if a primitive security separates states at the terminal date, then there exist multiperiod European options on that security generating dynamically complete markets. Second, we show that if a primitive security conditionally separates states at the terminal date, then there exist multiperiod European options on that security generating generically dynamically complete markets provided that certain conditions hold. Third, we show that there are economies for which the minimum number of multiperiod European options on a primitive security generating generically dynamically complete markets is relatively large. Finally, we show that in these economies, a relatively small number of multiperiod European options on possibly different portfolio strategies of primitive securities generates generically dynamically complete markets.     

Generalizations of Ho–Lee’s binomial interest rate model I: from one- to multi-factor

In this paper a multi-factor generalization of Ho–Lee model is proposed. In sharp contrast to the classical Ho–Lee, this generalization allows for those movements other than parallel shifts, while it still is described by a recombining tree, and is a process with stationary independent increments to be compatible with principal component analysis. Based on the model, generalizations of duration-based hedging are proposed. A continuous-time limit of the model is also discussed. This research was supported by Open Research Center Project for Private Universities: matching fund subsidy from MEXT, 2004–2008 and also by Grants-in-Aids for Scientific Research (No. 18540146) from the Japan Society for Promotion of Sciences.     

Use of minimum risk approach in the estimation of regression models with missing observations

This article considers a linear regression model with some missing observations on the response variable and presents two estimators of regression coefficients employing the approach of minimum risk estimation. Small disturbance asymptotic properties of these estimators along with the traditional unbiased estimator are analyzed and conditions, that are easy to check in practice, for the superiority of one estimator over the other are derived. Received May 2001     

Locally minimum variance unbiased estimator in a discontinuous density function

The exact forms of the locally minimum variance unbiased estimators and their variances are given in the case of a discontinuous density function.     

Sample Design and Allocation for Random Digit Dialling

Sample design and sample allocation methods are developed for random digit dialling in household telephone surveys. The proposed method is based on a two-way stratification of telephone numbers. A weighted probability proportional to size sample allocation technique is used, with auxiliary variables about the telephone coverage rates, within local telephone exchanges of each substrata. This makes the sampling design nearly “self-weighting” in residential numbers when the prior information is well assigned. A computer program generates random numbers for the local areas within the existing phone capacities. A simulation study has shown greater sample allocation gain by the weighted probabilities proportional to size measures over other sample allocation methods. The amount of dialling required to obtain the sample is less than for proportional allocation. A decrease is also observed on the gain in sample allocation for some methods through the increasing sample sizes.     

基于实物期权法对R&D项目价值评估的研究——对布莱克-舒尔斯定价模型的改进

针对R&D项目投资的特点,探讨了采用布莱克一舒尔斯期权定价模型对R&D项目价值评估可能存在的缺陷,并提出一种改进方法,即将决策树和布莱克一舒尔斯定价模型结合运用,因为决策树能够模拟研发项目的阶段性决策过程,考虑到多个离散型不确定性因素的相关性,模拟并计算出对决策路径依赖的现金流,因此能克服纯粹使用布莱克一舒尔斯公式的不足,在考虑多个不确定性因素的影响下,实现对多阶段R&D项目价值的评估,作出正确的投资决策。