The neglog transformation and quantile regression for the analysis of a large credit scoring database

Summary.  A statistical analysis of a bank's credit card database is presented. The database is a snapshot of accounts whose holders have missed a payment on a given month but who do not subsequently default. The variables on which there is information are observable measures on the account (such as profit and activity), and whether actions that are available to the bank (such as letters and telephone calls) have been taken. A primary objective for the bank is to gain insight into the effect that collections activity has on on-going account usage. A neglog transformation that highlights features that are hidden on the original scale and improves the joint distribution of the covariates is introduced. Quantile regression, a novel methodology to the credit scoring industry, is used as it is relatively assumption free, and it is suspected that different relationships may be manifest in different parts of the response distribution. The large size is handled by selecting relatively small subsamples for training and then building empirical distributions from repeated samples for validation. In the application to the database of clients who have missed a single payment a substantive finding is that the predictor of the median of the target variable contains different variables from those of the predictor of the 30% quantile. This suggests that different mechanisms may be at play in different parts of the distribution.     

The problem of defining nature first: A philosophical critique of environmental ethics

What is nature? Perhaps it is something of our own invention, perhaps not; but before we can begin researching the proper way to conserve it, we must come up with an accurate notion of what it is. Unless we examine the values we place on the environment and the resulting conflicts in which we trap ourselves, we will never be able to employ our definitions of nature, nor argue for our reactions toward it. Without a unified concept, perceptions of nature will create conflicting methodologies in terms of the environment—most of all, in terms of how we see it. Without a perception of the natural world, we aren't quite sure what it is we are trying to save. An accurate and focused perception of nature as a concept is the key. Until we understand the methodologies which influence our value placing systems, how we regard nature and how we develop ethics toward it will not be joined into a common consensus of practice.     

The information for the direction of dependence in l1 regression

Fitting a linear regression for a response variable by minimising the sum of absolute deviations, L₁ regression, may be viewed as a maximum likelihood procedure applied to the Laplace distribution. An interesting bivariate case is where the conditional distribution of the response X₂ given X₁ and the marginal distribution of the explanatory variable X₁ are both Laplace. In this context we show there is information to distinguish the direction of dependence between X₁ and X₂ from observations. That is we may distinguish the model in which X₁ is dependent on X₂ from that in which X₂ is dependent on X₁ This is not true for L₂ regression based on the Normal distribution.     

A Deliberative Method for Ranking Risks (II): Evaluation of Validity and Agreement among Risk Managers

A deliberative method for ranking risks was evaluated in a study involving 218 risk managers. Both holistic and multiattribute procedures were used to assess individual and group rankings of health and safety risks facing students at a fictitious middle school. Consistency between the rankings that emerged from these two procedures was reasonably high for individuals and for groups, suggesting that these procedures capture an underlying construct of riskiness. Participants reported high levels of satisfaction with their groups' decision-making processes and the resulting rankings, and these reports were corroborated by regression analyses. Risk rankings were similar across individuals and groups, even though individuals and groups did not always agree on the relative importance of risk attributes. Lower consistency between the risk rankings from the holistic and multiattribute procedures and lower agreement among individuals and groups regarding these rankings were observed for a set of high-variance risks. Nonetheless, the generally high levels of consistency, satisfaction, and agreement suggest that this deliberative method is capable of producing risk rankings that can serve as informative inputs to public risk-management decision making.     

Single subject research methods in rehabilitation

The article describes single-subject research designs as an efficient and cost-effective way to assess the impact of targeted interventions on individual behavior. The authors overview the basic tenets of single-subject research, provide examples of several types of single-subject designs, and describe the utility of single-subject designs in rehabilitation research.     

Standard errors for the parameters of graphical Gaussian models

The comparison of an estimated parameter to its standard error, the Wald test, is a well known procedure of classical statistics. Here we discuss its application to graphical Gaussian model selection. First we derive the Fisher information matrix and its inverse about the parameters of any graphical Gaussian model. Both the covariance matrix and its inverse are considered and a comparative analysis of the asymptotic behaviour of their maximum likelihood estimators (m.l.e.s) is carried out. Then we give an example of model selection based on the standard errors. The method is shown to produce almost identical inference to likelihood ratio methods in the example considered.     

The future of residential group care

The author addresses some of the tensions and strains that have affected the development of group care in the North American context. An argument is advanced for rethinking group care resources within a total continuum of child and family services.     

The role of family therapists in veterinary medicine: opportunities for clinical services, education, and research

Marriage and family therapists (MFTs) are applying their specific skill set in a variety of arenas. A new area for collaboration is veterinary medicine. The veterinary medical profession is emphasizing the importance of non-biomedical skills such as communication skills, acknowledging that human clientele are likely to view their pets as family members, and discussing veterinarian personal well-being. Each of these trends has clear application for intervention by MFTs. A discussion of how MFTs may be uniquely positioned to assist veterinary medicine is presented. An example of collaboration between MFT and veterinary medicine at Kansas State University is highlighted. Recommendations are made for development of effective educational relationships and possible private sector collaborations.