When Costs Are Unequal and Unknown: A Subtree Grafting Approach for Unbalanced Data Classification*

In binary classifications, a decision tree learned from unbalanced data typically creates an important challenge related to the high misclassification rate of the minority class. Assigning different misclassification costs can address this problem, though usually at the cost of accuracy for the majority class. This effect can be particularly hazardous if the costs cannot be specified precisely. When the costs are unknown or difficult to determine, decision makers may prefer a classifier with more balanced accuracy for both classes rather than a standard or cost‐sensitively learned one. In the context of learning trees, this research therefore proposes a new tree induction approach called subtree grafting (STG). On the basis of a real bank data set and several other data sets, we test the proposed STG method and find that our proposed approach provides a successful compromise between standard and cost‐sensitive trees.     

一个动态学生收费系统的设计

本文采用UCDOS希望汉字操作系统所提供的SDK开发工具在Fozpro数据库开发系统下，采用动态数据库的生成方法设计了一个学生收费管理系统，避免了传统静态数据库的缺陷。     

Detecting homogeneous segments in DNA sequences by using hidden Markov models

In recent years there has been a rapid growth in the amount of DNA being sequenced and in its availability through genetic databases. Statistical techniques which identify structure within these sequences can be of considerable assistance to molecular biologists particularly when they incorporate the discrete nature of changes caused by evolutionary processes. This paper focuses on the detection of homogeneous segments within heterogeneous DNA sequences. In particular, we study an intron from the chimpanzee α-fetoprotein gene; this protein plays an important role in the embryonic development of mammals. We present a Bayesian solution to this segmentation problem using a hidden Markov model implemented by Markov chain Monte Carlo methods. We consider the important practical problem of specifying informative prior knowledge about sequences of this type. Two Gibbs sampling algorithms are contrasted and the sensitivity of the analysis to the prior specification is investigated. Model selection and possible ways to overcome the label switching problem are also addressed. Our analysis of intron 7 identifies three distinct homogeneous segment types, two of which occur in more than one region, and one of which is reversible.     

Partial least squares estimator for single-index models

The partial least squares (PLS) approach first constructs new explanatory variables, known as factors (or components), which are linear combinations of available predictor variables. A small subset of these factors is then chosen and retained for prediction. We study the performance of PLS in estimating single-index models, especially when the predictor variables exhibit high collinearity. We show that PLS estimates are consistent up to a constant of proportionality. We present three simulation studies that compare the performance of PLS in estimating single-index models with that of sliced inverse regression (SIR). In the first two studies, we find that PLS performs better than SIR when collinearity exists. In the third study, we learn that PLS performs well even when there are multiple dependent variables, the link function is non-linear and the shape of the functional form is not known.     

Efficiency improvement from multiple perspectives: An application to Japanese banking industry

The current study focuses on efficiency improvement for banking systems from multiple perspectives, which have different definitions of input/output about various attributes of a banking system. In this research we utilize data envelopment analysis (DEA) and Nash bargaining game (NBG) theory to improve inefficient banks in order to: (1) Make the inefficient bank be the state of Pareto Optimality for multiple perspectives, which can avoid discontentment of some perspectives. (2) Improve a bank by changing its attributes and provide various improving schemes for decision makers. A numerical case study of Japanese banks is also given to show the results of equilibrium solution from multiple perspectives.     

隐含概念漂移的数据流分类算法研究

数据流具有快速性、连续性、多变性及无限性等特性，使得传统的分类方法不再适用。由于数据流多变的特性，设计高效、高精度的分类算法是比较困难的。目前数据流在电信、网络等众多应用领域客观大量存在，因此研究快速的、精确的、稳定的数据流分类系统具有较高的理论价值和应用前景。近年来，大量的研究工作都旨在高效准确地解决隐含概念漂移的数据流的分类问题。本文研究了一些适合数据流的分类算法，根据算法主要思想的不同进行分类，根据每种思想的发展历程对其相应的算法进行论述，并对部分经典算法分析其处理概念漂移的性能，并对进一步可做的工作进行展望。     

Bayesian Inference in Dynamic Disequilibrium Models: An Application to the Polish Credit Market

We propose a Bayesian approach for inference in a dynamic disequilibrium model. To circumvent the difficulties raised by the Maddala and Nelson (1974) specification in the dynamic case, we analyze a dynamic extended version of the disequilibrium model of Ginsburgh et al. (1980). We develop a Gibbs sampler based on the simulation of the missing observations. The feasibility of the approach is illustrated by an empirical analysis of the Polish credit market, for which we conduct a specification search using the posterior deviance criterion of Spiegelhalter et al. (2002).     

In-Sample or Out-of-Sample Tests of Predictability: Which One Should We Use?

It is widely known that significant in-sample evidence of predictability does not guarantee significant out-of-sample predictability. This is often interpreted as an indication that in-sample evidence is likely to be spurious and should be discounted. In this paper, we question this interpretation. Our analysis shows that neither data mining nor dynamic misspecification of the model under the null nor unmodelled structural change under the null are plausible explanations of the observed tendency of in-sample tests to reject the no-predictability null more often than out-of-sample tests. We provide an alternative explanation based on the higher power of in-sample tests of predictability in many situations. We conclude that results of in-sample tests of predictability will typically be more credible than results of out-of-sample tests.     

Testing for Circular Reflective Symmetry about a Known Median Axis

Circular data arise in many contexts, a particularly rich source being animal orientation experiments. Often, in the analysis of such data, a fundamental question of scientific interest is whether the underlying distribution is reflectively symmetric about some specific axis. In this paper, the situation in which the axis of interest is known to be a median axis is considered and a simple, asymptotically distribution- free test for circular reflective symmetry against skew alternatives is developed. The results from a simulation study lead to a testing strategy incorporating the new test and the circular analogue of the modified runs test of Modarres & Gastwirth (1996). The application of the testing strategy is illustrated using circular data arising from two animal orientation experiments.     

Production efficiency in Indian agriculture: An assessment of the post green revolution years

In this paper we use the nonparametric method of Data Envelopment Analysis (DEA) to obtain Pareto-Koopmans measures of technical efficiency of individual states over the years 1970–71 through 2000–01 in a multi-output, multi-input model of agricultural production. We disaggregate overall efficiency into two distinct components representing output and input efficiencies and identify the contribution of individual outputs and inputs to the measured level of overall efficiency. Because introduction of modern inputs has been a major component of the process of modernization of Indian agriculture, we examine to what extent different states succeeded in utilizing the modern inputs compared to the traditional inputs. Variations in the DEA efficiency scores across states and over years is explained in terms of differences in various institutional and demographic factors in a second stage regression analysis.