Multicollinearity and the value of a priori information

A measure of multicollinearity is defined which is useful in evaluating maintained hypotheses and aiding estimator selection as it suggests when a non-traditional estimator proposed by Bock (1975) is minimax and dominates ordinary least squares. An example is used to illustrate the presented methodology.     

A Methodology for Seismic Risk Analysis of Nuclear Power Plants

This methodology begins by quantifying the fragility of all key components and structures in the plant. By means of the logic encoded in the plant event trees and fault trees, the component fragilities are combined to form fragilities for the occurrence of plant damage states or release categories. Combining these, in turn, with the seismicity curves yields the frequencies of those states or releases. Uncertainty is explicitly included at each step of the process.     

Physician strives to create lean,clean health care machine. Studies of manufacturing processes may one day help make your practice more efficient

Elisabeth Hager, MD, MMM, CPE, is teaming up with scientists and industrialists to teach physicians how to apply principles of lean, total-quality manufacturing to their practices. She believes innovation and efficiencies can help doctors resurrect their profession's image and their control over it--and perhaps even reinvent American health care.     

A scope-of-practice survey leading to the development of standards of practice for health promotion in higher education

To review and analyze the scope of practice of health promotion services and draft standards of quality indicators for higher education communities, the American College Health Association (ACHA) initiated a Task Force on Health Promotion in Higher Education in May 1996. Members of the task force developed a National Survey on Health Promotion and Education in Institutions of Higher Education and mailed the survey to a stratified random sample of 600 ACHA member institutions, as well as to 97 key "best-practice health promotion leaders". The larger sample produced a 75.3% response rate, and 90.7% of the key informants returned usable surveys. The authors report selected findings from both groups that chronicle the state of health promotion practice in higher education at the close of the 20th century. The task force used the findings to establish a data-driven framework for the Year 2001 Standards of Practice for Health Promotion in Higher Education.     

The wireless network jamming problem

In adversarial environments, disabling the communication capabilities of the enemy is a high priority. We introduce the problem of determining the optimal number and locations for a set of jamming devices in order to neutralize a wireless communication network. This problem is known as the wireless network jamming problem. We develop several mathematical programming formulations based on covering the communication nodes and limiting the connectivity index of the nodes. Two case studies are presented comparing the formulations with the addition of various percentile constraints. Finally, directions of further research are addressed.     

Kernel estimation of a distribution function

A distribution function is estimated by a kernel method with

a poinrwise mean squared error criterion at a point x. Relation- ships between the mean squared error, the point x, the sample size and the required kernel smoothing parazeter are investigated for several distributions treated by Azzaiini (1981). In particular it is noted that at a centre of symmetry or near a mode of the distribution the kernei method breaks down. Point- wise estimation of a distribution function is motivated as a more useful technique than a reference range for preliminary medical diagnosis.     

Binge Drinking Among Undergraduate College Students in the United States: Implications for Other Substance Use

Abstract

The authors examined the relationship between binge drinking and other substance use among US college students, using nationally representative data from the 1995 National College Health Risk Behavior Survey implemented by the Centers for Disease Control and Prevention. Compared with nonbinge drinkers, current binge drinkers were significantly more likely to report “ever” using and current use of cigarettes, marijuana, cocaine, and other illegal drugs. The researchers also found that the more often students binge drank, the more likely they were to have ever used cigarettes, marijuana, cocaine, and other drugs, and the more likely they were to report current use of cigarettes and marijuana. Those who design programs to prevent binge drinking and use of other substances should take into account the reality that many students use more than one substance and that the more frequently students report binge drinking, the more likely they are to be using other substances as well.