Variable reduction,sample selection bias and bank retail credit scoring

This paper investigates the effect of including the customer loan approval process to the estimation of loan performance and explores the influence of sample selection bias in predicting the probability of default. The bootstrap variable reduction technique is applied to reduce the variable dimension for a large data-set drawn from a major UK retail bank. The results show a statistically significant correlation between the loan approval and performance processes. We further demonstrate an economically significant improvement in forecasting performance when taking into account sample selection bias. We conclude that financial institutions can obtain benefits by correcting for sample selection bias in their credit scoring models.     

Bimodal issues,the median voter model,Legislator's ideology,and abortion

The median voter model is widely used in the public choice literature to explain legislator's behavior. According to the model, if voter preferences are unimodal, a vote-maximizing legislator should mirror the position of the median voter. However, the median voter model has not been tested on bimodal issues. This paper fills this critical void by empirically testing the applicability of the median voter model on an issue which clearly meets the criteria for being bimodal: abortion. Using a variety of attitudinal measures from large sample public opinion polls and constituency demographics, this study finds that Senate voting on the 1994 Freedom of Abortion Access bill was highly related to the senator's personal characteristics—especially ideology—and not to constituent opinion or demographics.     

Contextual marketing--the real business of the Internet

The painful truth is that the Internet has been a letdown for most companies--largely because the dominant model for Internet commerce, the destination Web site, doesn't suit the needs of those companies or their customers. Most consumer product companies don't provide enough value or dynamic information to induce customers to make the repeat visits--and disclose the detailed information--that make such sites profitable. In this article, David Kenny and John F. Marshall suggest that companies discard the notion that a Web site equals an Internet strategy. Instead of trying to create destinations that people will come to, companies need to use the power and reach of the Internet to deliver tailored messages and information to customers. Companies have to become what the authors call "contextual marketers." Delivering the most relevant information possible to consumers in the most timely manner possible will become feasible, the authors say, as access moves beyond the PC to shopping malls, retail stores, airports, bus stations, and even cars. The authors describe how the ubiquitous Internet will hasten the demise of the destination Web site--and open up scads of opportunities to reach customers through marketing "mobilemediaries," such as smart cards, e-wallets, and bar code scanners. The companies that master the complexity of the ubiquitous Internet will gain significant advantages: they'll gain greater intimacy with customers and target market segments more efficiently. The ones that don't will be dismissed as nuisances, the authors conclude. They suggest ways to become welcome additions--not unwelcome intrusions--to customers' lives.     

沃顿商学院中国的第二次经济大转型

我一直密切关注着中国近年来增长与繁荣。现在是中国的第二次经济转型．即构筑统一的国内市场的时候了。我认为这一转变势在必行．只有如此．那些有能力在国内竞争中获胜，继而挺进国际市场的企业才能脱颖而出。     

What Is Personalization? Perspectives on the Design and Implementation of Personalization in Information Systems

In e-commerce and mobile commerce, personalization has been recognized as an important approach element in customer relationships and Web strategies. However, there are wide differences in how this concept is defined, characterized, and implemented in the literature. In this article we present a high-level framework for classifying approaches to personalization that delineates fundamental assumptions about personalization in the literature and relates them to strategies for developing personalization systems. The framework consists of 2 parts: (a) a set of perspectives on personalization that guide the design of personalization systems at a general level and (b) a scheme for classifying how personalization can be implemented. The personalization perspectives represent 4 distinct schools of thought on the nature of personalization distilled from the literature of several fields. These perspectives are ideal types and we discuss them in terms of the motivation they supply for personalization, the goals and means of personalization, and the ways in which they conceptualize and model users. The implementation classification scheme is constructed on 3 dimensions of implementation choices. These 3 dimensions pertain to what to personalize (content, interface, functionality, channel), to whom to personalize (individuals or categories of individuals) as well as who does the personalization (implicit or explicit personalization). The personalization perspectives represent particular concepts of personalization that guide general design choices; these choices are implemented via the options described in the implementation classification scheme. The framework contributes to the development of a common theoretical basis for the study of personalization. We discuss implications of the framework for design of personalization systems and future research directions.     

Studying (Small) businesses with the Michigan Employment Security Commission longitudinal data base

This paper addresses the usefulness of a longitudinal data file constructed from records on employers from the Michigan Employment Security Commission. We describe the main features of the data file, which includes quarterly (and in some cases, monthly) data from the third quarter of 1978 through the first quarter of 1983, plus the fourth quarters of 1983–87. We then illustrate the uses of the data with two examples: (1) studying changes in the Michigan economy, in particular the early growth and survival of new units of different sizes; and (2) studying the behavior of wages and employment following changes in ownership.Judith Connor and Steven Heeringa are primarily responsible for Section II and the Appendix, John Jackson for Section III, and Charles Brown for Section IV.     

Bank Capital Standards for Market Risk: A Welfare Analysis

We propose a simple model that is suitable for evaluating alternative bank capital regulatory proposals for market risk. Our model formalizes the conflict between bank objectives and regulatory goals. Banks' decisions represent a tension between their desire to exploit the deposit-insurance put option and their desire to preserve franchise value. Regulators seek to balance the social value of deposits in mediating transactions against the deadweight costs of failure resolution. Our social welfare criterion is standard: a weighted average agents' utilities.We demonstrate that banks do not incrementally alter their portfolio risk as the economic environment changes. Rather, banks either choose the minimal feasible risk or the maximal feasible risk. This pattern, in turn, drives regulatory decisions: The first goal of the regulator is to induce banks to choose the minimal risk level. For all nontrivial cases, unregulated banks fail to choose the first-best allocations. Traditional ex-ante capital requirements can induce banks to choose the socially-optimal level of portfolio risk, but the required capital is often inefficiently high. In contrast, variants of the Federal Reserve Board's precommitment proposal imply far smaller efficiency losses, and achieve allocations at or near the first-best for most reasonable model specifications. The ex-post penalties required for the optimal implementation of precommitment are not excessively large. The welfare gains from precommitment are even higher when the precommitment penalty function is precluded from sending banks into default. We conclude that state-contingent regulatory mechanisms, of which the precommitment approach is an example, offer the possibility of substantial gains in regulatory efficiency, relative to traditional state non-contingent regulation.     

Wage Determination Regimes and Pay Inequality: A comparative study of Latin American countries

The relationship between the degree of centralization of wage setting and labour market outcomes, generally discussed with reference to advanced OECD countries, is explored in this paper through the comparative analysis of eight Latin American countries, focusing on the effects of centralization on wage inequality. It is argued that the greater the decentralization of wage setting, the larger will be wage dispersion. The article starts with the presentation of the analytical model, and continues with the empirical study. This is divided into two parts. In the first, a number of institutional factors that contribute to define 'wage determination regimes' (collective bargaining structures; state intervention in wage setting; and union rights that determine union strength) are discussed, stressing the specific traits of centralization within Latin America, and the selected countries are scored in terms of those factors. In the second, the magnitude of wage dispersion across manufacturing industries is examined, and the resulting ranking of countries is contrasted with their ranking in terms of the institutional factors. It is concluded that, with some caveats, the degrees of centralization of bargaining structures and of wage inequality appear to be inversely associated, reproducing in the Latin American region the pattern found in OECD countries.