This article introduces to the statistical and insurance literature a mathematical technique for an a priori classification of objects when no training sample exists for which the exact correct group membership is known. The article also provides an example of the empirical application of the methodology to fraud detection for bodily injury claims in automobile insurance. With this technique, principal component analysis of RIDIT scores (PRIDIT), an insurance fraud detector can reduce uncertainty and increase the chances of targeting the appropriate claims so that an organization will be more likely to allocate investigative resources efficiently to uncover insurance fraud. In addition, other (exogenous) empirical models can be validated relative to the PRIDIT-derived weights for optimal ranking of fraud/nonfraud claims and/or profiling. The technique at once gives measures of the individual fraud indicator variables' worth and a measure of individual claim file suspicion level for the entire claim file that can be used to cogently direct further fraud investigation resources. Moreover, the technique does so at a lower cost than utilizing human insurance investigators, or insurance adjusters, but with similar outcomes. More generally, this technique is applicable to other commonly encountered managerial settings in which a large number of assignment decisions are made subjectively based on "clues," which may change dramatically over time. This article explores the application of these techniques to injury insurance claims for automobile bodily injury in detail.

Fraud Classification Using Principal Component Analysis of RIDITs

Child survival in big cities: the disadvantages of migrants.

Analysis of the 1986 Senegal Demographic and Health Survey reveals that mothers may be able to improve their children’s survival chances by migrating from the countryside to the city. Children of urban migrants, however, continue to experience a much higher risk of mortality before the age of 5 than children of urban nonmigrants, even after the mother has lived in the city for several years. This migrant mortality disadvantage persists when controlling for numerous socioeconomic and fertility-related factors typically associated with child mortality in developing countries, which also serve as indicators of migrant selection and adaptation.

Rural-to-Urban migration and child survival in Senegal

Evidence regarding the effects of health status on migration are mixed with some concluding that there is a positive relation while others conclude a negative relation. This study examines the analytical bases for conflicting findings and proposes a model of the health-migration relation which allows for age selectivity nonlinearities interactions and dynamic effects. Often in migration research we focus on migrant-nonmigrant differentials which helps us to understand the cross-sectional differences between the 2 groups of persons but it does little to elucidate the complex process by which nonmigrants become migrants. The sequence of migration and health events from the US National Health Interview Survey from 1979-1980 show that for the population under age 45 there is little temporal connection between migration and health status changes. 2/3 of the younger population with at least 1 health or migration event migrated only with health status changes neither preceding nor following the migration. Since the migration and health events for these persons are not associated in time there is little basis for positing strong relations of either direction for the nonelderly. The age groups for which migration and health events are connected in time are those over age 45 and the temporal proximity becomes even more marked for the elderly. Among the elderly the migrant-only group constituted only 10% of those with at least 1 event. As expected for the elderly the dominant relation is health deterioration after migration the pattern displayed by 65% of the elderly with events in the last 5 years. The elderly are also more likely to be in the Destabilized Migrant category experiencing health status declines both before and after migration. This is also consistent with the theory that the elderly in poor health at the time of migration will be more likely to be stressed by their migration and experience further deteriorations in health. The conditional probabilities indicate that health status declines are unlikely to precipitate immediate migrations except for those with pre-existing serious conditions or more severe activity limitations. (authors modified)

The directionality and age selectivity of the health-migration relation: evidence from sequences of disability and mobility in the United States.

Background Canine rabies was reintroduced to the city of Arequipa, Peru in March 2015. The Ministry of Health has conducted a series of mass dog vaccination campaigns to contain the outbreak, but canine rabies virus transmission continues in Arequipa's complex urban environment, putting the city's 1 million inhabitants at risk of infection. The proximate driver of canine rabies in Arequipa is low dog vaccination coverage. Our objectives were to qualitatively assess barriers to and facilitators of rabies vaccination during mass campaigns, and to explore strategies to increase participation in future efforts. Methodology/principal findings We conducted 8 focus groups (FG) in urban and peri-urban communities of Mariano Melgar district; each FG included both sexes, and campaign participants and non-participants. All FG were transcribed and then coded independently by two coders. Results were summarized using the Social Ecological Model. At the individual level, participants described not knowing enough about rabies and vaccination campaigns, mistrusting the campaign, and being unable to handle their dogs, particularly in peri-urban vs. urban areas. At the interpersonal level, we detected some social pressure to vaccinate dogs, as well as some disparaging of those who invest time and money in pet dogs. At the organizational level, participants found the campaign information to be insufficient and ill-timed, and campaign locations and personnel inadequate. At the community level, the influence of landscape and topography on accessibility to vaccination points was reported differently between participants from the urban and peri-urban areas. Poor security and impermanent housing materials in the peri-urban areas also drives higher prevalence of guard dog ownership for home protection; these dogs usually roam freely on the streets and are more difficult to handle and bring to the vaccination points. Conclusions A well-designed communication campaign could improve knowledge about canine rabies. Timely messages on where and when vaccination is occurring could increase dog owners' perception of their own ability to bring their dogs to the vaccination points and be part of the campaign. Small changes in the implementation of the campaign at the vaccination points could increase the public's trust and motivation. Location of vaccination points should take into account landscape and community concerns.

/pdf/barriers-to-dog-rabies-vaccination-during-an-urban-rabies-29evnf3r9m.pdf

G. Diebold

Papers

Health effects of rural-urban migration in developing countries—Senegal☆