Evaluating Expert Systems in a Financial Domain: A Study Demonstrating Issues in Case Selection,Validation Standards and Evaluator Bias

This article describes and generalizes a validation study of four commercially available personal financial planning expert systems and the rationale for the research methodology used. Our evaluation of these systems adds to the understanding of verification and validation issues related to case selection, validation standards and evaluator bias. The article describes the systems, their domain and the empirical method—field tests using hypothetical cases—and relates that method to the literature. Comparing same-task systems combines multiple system perspectives and multiple models. Our methodology did efficiently and effectively identify conflicting terminology, omissions and system weaknesses but was inadequate for comparing the complex plan recommendations. The results re-emphasize the importance of continual knowledge base updating, formal system testing and the need for external evaluation. The results also show the value of comparing multiple, same-task systems.     

Bias in estimates of discrimination and default in mortgage lending: The effects of simultaneity and self-selection

The common practices of estimating single-equation models of mortgage rejection to test for discrimination in mortgage markets or single-equation ex ante mortgage default equations to validate underwriting criteria produce biased and inconsistent parameter estimates. This is due to problems of simultaneous equations bias which arise because, in a world of imperfect information, mortgage terms are not exogenous to the rejection or default decision. In addition, mortgage default estimates are also subject to selection bias. Monte Carlo experiments are used to study the nature and extent of likely bias in single-equation estimation results. We find that rejection equation estimates indicate discrimination when none exists and that estimated coefficients of mortgage terms, such as the loan-to-value ratio, are also subject to significant bias in both rejection and default equations.     

Medical Error Disclosure Training: Evidence for Values-Based Ethical Environments

Disclosure of medical and errors to patients has been increasingly mandated in the U.S. and Canada. Thus, some health systems are developing formal disclosure policies. The present study examines how disclosure training may impact staff and the organization. We argue that organizations that support “disclose and apologize” activities, as opposed to “deny and defend,” are demonstrating values-based ethics. Specifically, we hypothesized that when health care clinicians are trained and supported in error disclosure, this may signal a values-based ethical environment, and staff may be more committed to the organization. We surveyed 325 clinical care providers employed by a large hospital that had recently begun implementing disclosure policies and training. Disclosure training explained significant variance in perceptions of the ethical environment, and the ethical environment mediated the relationship between disclosure training and organizational commitment. Although this study explored disclosure of medical errors, organizational support for error disclosure is a concept that could be relevant for many types of organizations.     

Should we let employees contract away their rights against arbitrary discharge?

This article argues that the moral right to be discharged only for good cause and like rights can be contracted away by employees in appropriate circumstances. It maintains that the rights in question are not inalienable, and that there is nothing irrational about an employee's wishing to deal them away. It also maintains that inequalities in bargaining power between employers and employees are insufficiently pervasive to justify a flat ban on the alienation of these rights. For a waiver of such rights to be valid, however, employees must have full knowledge of its terms.The question addressed here bears on several legal and policy issues affecting termination of the employment relation. If employees can contract away their right to a goodcause discharge, the American doctrine of employment at willmight find justification in the face of that right. In addition, the alienability of such discharge rights may be necessary to justify express disclaimers of wrongful discharge liability and the waiver provision of the new U.S. Draft Uniform Employment-Termination Act.Michael J. Phillips is Professor of Business Law and Chairman of the Business Law Department of Indiana University's School of Business. He is a former editor-in-chief of theAmerican Business Law Journal, and has authored numerous law journal articles.     

Estimating the Cost of Equity Capital for Property-Liability Insurers

This article presents new evidence on the cost of equity capital by line of insurance for the property‐liability insurance industry. To do so we obtain firm beta estimates and then use the full‐information industry beta (FIB) methodology to decompose the cost of capital by line. We obtain full‐information beta estimates using the standard one‐factor capital asset pricing model and extend the FIB methodology to incorporate the Fama–French three‐factor cost of capital model. The analysis suggests the cost of capital for insurers using the Fama–French model is significantly higher than the estimates based upon the CAPM. In addition, we find evidence of significant differences in the cost of equity capital across lines.     

The Market for Corporate Assets: Who Engages in Mergers and Asset Sales and Are There Efficiency Gains?

We analyze the market for corporate assets. There is an active market for corporate assets, with close to seven percent of plants changing ownership annually through mergers, acquisitions, and asset sales in peak expansion years. The probability of asset sales and whole-firm transactions is related to firm organization and ex ante efficiency of buyers and sellers. The timing of sales and the pattern of efficiency gains suggests that the transactions that occur, especially through asset sales of plants and divisions, tend to improve the allocation of resources and are consistent with a simple neoclassical model of profit maximizing by firms.     

A sequential approach to testing for structural change in econometric models

The paper shows that the sequential approach to testing econometric models, particularly testing for structural change, is both feasible and potentially very useful. In fact, this paper makes clear the possibility of using the sequential approach as suggested by Dhrymes et al. (1972) and shows that the statistical dependence between successive tests can be overcome in some cases.Helpful comments by David Hendry, Grayham Mizon and Jan Kiviet, on an earlier version of a related Paper, are gratefully acknowledged.