Neural networks and flexible regression and discrimination

Feed-forward neural networks—also known as multi-layer perceptrons—are now widely used for regression and classification. In parallel but slightly earlier, a family of methods for flexible regression and discrimination were developed in multivariate statistics, and tree-induction methods have been developed in both machine learning and statistics. We expound and compare these approaches in the context of a number of examples.     

An economic analysis of public sector collective bargaining and strike activity

This paper develops a theory of public sector collective bargaining and uses it to investigate the economic determinants of public sector strike activity. The model considers union leaders, union members, bureaucrats, politicians, and voters, with the intention of explicitly recognizing both the constraints placed on the bargaining parties by their constituents and the differences between public and private sector collective bargaining. The empirical results indicate that only in cases where both negotiating parties are motivated by self-interest can we expect frequent strikes and strikes of long duration. The results also suggest that public sector strikes are countercyclical and principally influenced by the business cycle and its impact on state and local revenues.     

An empirical excursion into the quicksand of vehicle replacement

The rapid increase of physical distribution costs relative to other categories has resulted in a renewed interest in the analysis of transport operations. One of the most common problems is that of vehicle replacement, and in this paper we present an analysis of real data. The methodology is developed paying due regard to the role of quantitative analysis in the decision-making process, and the sort of data which is readily available in most transport departments. A traditional discounted cost model is presented and this is followed by a productivity-based analysis, which is shown to be an improvement. The latter generates optimal solutions and is flexible enough to carry out considerable sensitivity analysis.     

Cost containment in health care: a model for management research.

Cost containment is a dominant problem in the health care field, but it has not been addressed from a comprehensive management perspective. To fill this gap, we have developed an inclusive model of the cost containment process. The model has implications for management research in several areas: cost containment baselines, incentive systems, organization structures, cost/quality trade-offs, and cost containment constraints.     

An Endangered Species: The Female as Addict or Member of an Addict Family*

When a woman is the identified addict or a member of a drug/alcohol abusing family system she often experiences considerably more psychiatric abuse than her male cohorts. This paper surveys some of the representative literature reflecting attitudes and practices imposed on these women. Explanations based on a study of healthy family systems are given with interpretations anchored in cultural rather than sexist phenomena. It is suggested that women need to be understood according to the complex interlocking transactions within family systems.     

Sonet-1: Social network analysis and modeling system

SONET-I is a preliminary but completely operational version of a package of computer programs for the analysis of social structure. The programs were written to implement a strategy for the analysis of social networks which the authors have developed over the past several years. The conceptual basis for the research method has its origins in graph theory on the one hand and in anthropological studies of social organization — especially kinship studies — on the other. The method can, in fact, be seen as a graph theoretic formalization and generalization of kinship-based methods of social analysis often associated with British social anthropology.     

On aliasing and generalized defining relationships of factorial arrangements

The concepts of defining contrast (DC), generalized defining relationship (GDR) and aliasing structure (AS) are now well established in the terminology of regression analysis and factorial design theory. There is no complete agreement in the literature about the meaning of regular and irregular fractional factorial designs. This paper provides a workable definition of a regular fraction from a symmetrial prime-powered factorial. It characterizes the uniqueness of the GDR for fractions from the most general factorial. Results are also présentés on the uniqueness of the GDR for regular designs, on orthogonality aspects of regular and irregular designs, and on group-theoretic generation of the complete aliasing structure. Examples are provided to illustrate the developments.