Evaluating Probabilistic Forecasts with Bayesian Signal Detection Models

We propose the use of signal detection theory (SDT) to evaluate the performance of both probabilistic forecasting systems and individual forecasters. The main advantage of SDT is that it provides a principled way to distinguish the response from system diagnosticity, which is defined as the ability to distinguish events that occur from those that do not. There are two challenges in applying SDT to probabilistic forecasts. First, the SDT model must handle judged probabilities rather than the conventional binary decisions. Second, the model must be able to operate in the presence of sparse data generated within the context of human forecasting systems. Our approach is to specify a model of how individual forecasts are generated from underlying representations and use Bayesian inference to estimate the underlying latent parameters. Given our estimate of the underlying representations, features of the classic SDT model, such as the receiver operating characteristic (ROC) curve and the area under the ROC curve (AUC), follow immediately. We show how our approach allows ROC curves and AUCs to be applied to individuals within a group of forecasters, estimated as a function of time, and extended to measure differences in forecastability across different domains. Among the advantages of this method is that it depends only on the ordinal properties of the probabilistic forecasts. We conclude with a brief discussion of how this approach might facilitate decision making.     

Measuring living standards with proxy variables

Very few demographic surveys in developing countries have gathered information on household incomes or consumption expenditures. Researchers interested in living standards therefore have had little alternative but to rely on simple proxy indicators. The properties of these proxies have not been analyzed systematically. We ask what hypotheses can be tested using proxies, and compare these indicators with consumption expenditures per adult, our preferred measure of living standards. We find that the proxies employed in much demographic research are very weak predictors of consumption per adult. Nevertheless, hypothesis tests based on proxies are likely to be powerful enough to warrant consideration.     

The UK National Telephone Gambling Helpline-Results on the First Year of Operation

This study outlines the results of the UK's national gambling helpline run by GamCare. The results outlined here cover the period of the first 12 months of operation (November 1997 to October 1998). The helpline received a total of 1729 calls. Of these, 51% were from problem gamblers themselves (90% male; 10% female) and a further 26% of calls were from relatives of problem gamblers. The remaining calls came from other professionals handling problem gambling cases (13%), attempted calls, e.g., people calling and then putting the phone down due to being scared of talking (4%), information requests (3%) and the media (3%). Fruit machine gambling appeared to be most problematic for the callers as a whole and for particular sub-groups such as adolescents (82%) and women (52%).     

Racial inequality in active life among adult americans

Is a shorter life with more years lived in poor health a defining attribute of the life cycle of disadvantaged groups? Based on the J990 5% Public Use Microdata Survey, we develop life table models of healthy (or active) life for the major racial groups, by sex, in the United States. The analysis underscores the complexity of the relationship between morbidity and mortality in the population. For Asians, longer life is associated with fewer years lived in poor health. In contrast, Native Americans’ relatively longer lives are accompanied by extended periods of chronic health problems. of all racial groups, blacks live the fewest years, and they live a high proportion of those years with a chronic health problem. Hispanics also live substantially fewer years, yet the period of life they spend with a health problem is relatively compressed. Racial differences in the link between morbidity and mortality point to the importance of investigating how chronic diseases and disease prevention and treatment are related to active life across the population subgroups.     

Terrorist Population Dynamics Model

A system that includes a number of terrorist cells is considered. The cells can consist of one or more terrorists. The current number of terrorist cells is further denoted by N(t), where t is a current time counted from any appropriate origin. The objective is to find the evolution of the system in terms of N(t) and some interpretable parameters, such as the initial number of the terrorist cells N0=N(0), the cell disabling rate constant lambda (or the cell half-life t1/2), and the rate of formation of new cells P. The cost-effectiveness analysis, performed in the framework of the model, reveals that the effectiveness of disabling a terrorist cell is getting worse after 2-3 half-lives of a cell, which shows that if the anti-terrorist actions have not reached their goal during that time, the respective policy should be considered for revision, using the risk assessment consideration. Another important issue raised concerns balancing the efforts related to counterterrorism actions inside the system and the efforts protecting its borders. The respective data analysis is suggested and illustrated using simulated data.     

Colleges respond to student binge drinking: reducing student demand or limiting access

Administrators at 68% of 4-year colleges nationwide (N = 747) responded to a survey concerning the types of programs and policies they used in response to students' heavy drinking. Most schools conducted targeted alcohol education and invested in institutional prevention efforts; half conducted social norms campaigns; a sizeable minority restricted alcohol on campus. Schools that focused on demand reduction were less likely to ban alcohol use. One in 3 schools received funding for these programs from governmental agencies, and 1 in 5 from the alcohol industry. Such schools were more likely to conduct targeted alcohol education and social norms programs and were less likely to restrict alcohol use on campus or at college events. Colleges may want to reconsider prevention initiatives that focus exclusively on demand or supply. They may also want to examine the extent to which funding is the driving force shaping the direction of their alcohol initiatives.