Search

Housing discrimination and racial segregation have a long history in the United States. The 1930s Home Owners' Loan Corporation (HOLC) "residential security maps," recently digitized, have become a popular visualization of Depression era mortgage lending risk patterns across American cities. Numerous housing policies have since been instituted, including the Home Mortgage Disclosure Act (HMDA), but mortgage lending bias persists. The degree to which detailed spatial patterns of bias have persisted or changed along with urban change is not well understood. We compare historic HOLC grades and contemporary levels of mortgage lending bias using spatially detailed HMDA data. We further examine the relationship between HOLC risk grades and contemporary racial and ethnic settlement patterns. Results suggest that historical mortgage lending risk categorizations and settlement patterns are associated with contemporary mortgage lending bias and racial and ethnic settlement patterns. Concerted and deliberate efforts will be needed to change these patterns. [ABSTRACT FROM AUTHOR]

The objective of this study was to examine the role of body mass and subcutaneous fat in injury severity and pattern sustained by overweight drivers. Finite element models were created to represent the geometry and properties of subcutaneous adipose tissue in the torso with data obtained from reconstructed magnetic resonance imaging data-sets. The torso adipose tissue models were then integrated into the standard multibody dummy models together with increased inertial parameters and sizes of the limbs to represent overweight occupants. Frontal crash simulations were carried out considering a variety of occupant restraint systems and regional body injuries were measured. The results revealed that differences in body mass and fat distribution have an impact on injury severity and pattern. Even though the torso adipose tissue of overweight subjects contributed to reduce abdominal injury, the momentum effect of a greater body mass of overweight subjects was more dominant over the cushion effect of the adipose tissue, increasing risk of other regional body injuries except abdomen. Through statistical analysis of the results, strong correlations (p < 0.01) were found between body mass index and regional body injuries except neck injury. The analysis also revealed that a greater momentum of overweight males leads to greater forward torso and pelvic excursions that account for higher risks (p < 0.001) of head, thorax and lower extremity injury than observed in non-overweight males. The findings have important implications for improving the vehicle and occupant safety systems designed for the increasing global obese population. [ABSTRACT FROM AUTHOR]

Here we extend predictive method for model selection of Laud and Ibrahim to the generalized linear model. This prescription avoids the need to directly specify prior probabilities of models and prior densities for the parameters. Instead, a prior prediction for the response induces the required priors. We propose normal and conjugate priors for generalized linear models, each using a single prior prediction for the mean response to induce suitable priors for each variable-subset model. In this way, an informative prior is used to select a subset of variables. In addition to producing a ranking of models by size of the predictive criterion, the standard deviation of the criterion is used as a calibration number to produce a set of equally good models. A straightforward Markov chain Monte Carlo algorithm is used to accomplish the necessary computations. We illustrate this method with real and simulated datasets and compare results with the Bayes factors and the Akaike information and Bayes information model selection criteria. The simulation results confirm the efficacy of the method. because the correct model is known. An illustrative application demonstrates selection of important predictors of success in identifying the sentinel lymph node during surgical treatment ot breast cancer. A forward selection procedure is described to avoid a full search over the 218 possible models in this case. [ABSTRACT FROM AUTHOR]

Presents a paper on the computation of several number theoretic functions using the application of probability. Blocks of numbers; Sum of ratio of blocks to bits.

Introduces the use of a predictive Bayesian criterion in the context of model selection problem based on the predictive density of a replicate experiment (PDRE). Provision of a calibration of the criterion; Relationships of the proposed criterion to other prevalent criteria; Illustration of the proposed methodology.