Robust regression through robust covariances

This paper discusses the estimation of regression parameters after summarizing the data by a covariance matrix of the concatenated vector of explanatory variables and response variable. A robust estimate of the covariance matrix leads to a robust regression estimator. An M-estimator at the covariance estimation step is studied in the paper, and the resulting regression estimator is compared to a few previously proposed robust regression estimators.     

Redescending M-estimators

In finite sample studies redescending M-estimators outperform bounded M-estimators (see for example, Andrews et al. [1972. Robust Estimates of Location. Princeton University Press, Princeton]). Even though redescenders arise naturally out of the maximum likelihood approach if one uses very heavy-tailed models, the commonly used redescenders have been derived from purely heuristic considerations. Using a recent approach proposed by Shurygin, we study the optimality of redescending M-estimators. We show that redescending M-estimator can be designed by applying a global minimax criterion to locally robust estimators, namely maximizing over a class of densities the minimum variance sensitivity over a class of estimators. As a particular result, we prove that Smith's estimator, which is a compromise between Huber's skipped mean and Tukey's biweight, provides a guaranteed level of an estimator's variance sensitivity over the class of densities with a bounded variance.     

A Symposium on Unemployment Statistics

Separate Poisson distributions were evaluated via several hypotheses tests for each of the 16 teams in the National Hockey League using 1973–74 goals for and goals against both home and away. After examining independence of goals for and goals against, they in turn were used to explain order of finish for the 16 teams. The results were quite surprising.     

Enhancing Self‐Protective Behavior: Efficacy Beliefs and Peer Feedback in Risk Communication

In times of a high‐impact safety incident citizens may have a variety of sources available to help them cope with the situation. This research focuses on the interplay of efficacy information in risk communication messages and peer feedback, such as responses on social network sites (SNSs) in the context of a high‐impact risk on the intention to engage in self‐protective behavior. The study pitted high and low efficacy information messages against supporting and opposing peer feedback (N = 242). Results show a significant interaction effect between efficacy information in a news article and peer feedback from SNS messages on both the intention to engage in self‐protective behavior and levels of involvement. Participants who received the article with more efficacy information and also received supportive peer feedback via SNS messages were more likely to express higher levels of involvement and greater intentions to engage in protective behavior. When confronted with a low efficacious news article, the effect of peer feedback on these two variables was significantly stronger. Finally, implications for theory and government risk communication are discussed.     

Maternal Hurricane Exposure and Fetal Distress Risk

Logistic regression and spatial analytic techniques are used to model fetal distress risk as a function of maternal exposure to Hurricane Andrew. First, monthly time series compare the proportion of infants born distressed in hurricane affected and unaffected areas. Second, resident births are analyzed in Miami‐Dade and Broward counties, before, during, and after Hurricane Andrew. Third, resident births are analyzed in all Florida locales with 100,000 or more persons, comparing exposed and unexposed gravid females. Fourth, resident births are analyzed along Hurricane Andrew's path from southern Florida to northeast Mississippi. Results show that fetal distress risk increases significantly with maternal exposure to Hurricane Andrew in second and third trimesters, adjusting for known risk factors. Distress risk also correlates with the destructive path of Hurricane Andrew, with higher incidences of fetal distress found in areas of highest exposure intensity. Hurricane exposed African‐American mothers were more likely to birth distressed infants. The policy implications of in utero costs of natural disaster exposure are discussed.     

Legitimacy of tolerating limited environmental pollution? The case for natural attenuation

Degradations of environmental quality often pose severe harms or at least adverse effects to individuals and societies. Perceiving any environmental pollution thus appears to be connected with an implicit claim for its prompt and complete removal. For example, the oil-spill from the "Prestige" accident at the western Spanish shoreline is surely still in everyone's mind, where—in a somewhat Sysiphos-effort—a lot of helpers tried to remove the huge masses of oil mud, washed repeatedly ashore. However, alternative rehabilitation conceptions are also conceivable for certain pollution problems, which may be less unhealthy and possibly more efficient. Among them, taking advantage of natural occurring processes, which are grouped in this context as "natural attenuation" (NA), is an emerging and challenging but possibly also a questionable approach. NA as remediation strategy is, therefore, often discussed controversially among experts and actors as well as in the public (Teutsch/Rügner 2000). This is partly due to still open scientific or technical questions but also due to legitimacy problems of natural attenuation on ethical and juridical levels (Steffens et al. 2002; Heinz 2002). Therefore, a balanced view on adequate natural attenuation seems to be necessary. The following analysis will review NA as rehabilitation means from the normative perspective, giving some general statements—and thus orientation—for public policy making as well as for the administrative level.