Tree-structured subgroup analysis for censored survival data: Validation of computationally inexpensive model selection criteria

The performance of computationally inexpensive model selection criteria in the context of tree-structured subgroup analysis is investigated. It is shown through simulation that no single model selection criterion exhibits a uniformly superior performance over a wide range of scenarios. Therefore, a two-stage approach for model selection is proposed and shown to perform satisfactorily. Applied example of subgroup analysis is presented. Problems associated with tree-structured subgroup analysis are discussed and practical solutions are suggested.     

An Evaluation of Hispanic Population Estimates*

Objective. Estimates of the Hispanic population have traditionally been based on historical trends, ratios, or some variant of the cohort‐component method. In this article, we describe and test a methodology in which estimates of the Hispanic population are based on symptomatic indicators of population change such as births, deaths, and school enrollments. Methods. Using a variety of techniques, we develop Hispanic population estimates for counties in Florida. We evaluate the accuracy of those estimates by comparing them with 2000 Census counts. Results. Hispanic population estimates have larger errors than estimates of total population; errors vary considerably by population size and growth rate; some techniques perform better than others in places with particular population characteristics; and averages often perform better than individual techniques. Conclusions. In many circumstances, symptomatic data series can provide more accurate estimates of the Hispanic population than more commonly used techniques.     

The Couch Project

Carl Couch had a distinctive project. He was a man of distinctive character. In this commentary, I first describe the Couch project and then discuss how the project was interwoven with the character of its author. I then discuss “Oh, What Webs Those Phantoms Spin” in the context of Carl's larger project and close with a brief discussion of New Iowa School research after Carl Couch.     

Modeling Social Processes: Some Lessons from Sports

Sports events can provide helpful models for thinking about social processes. DiMaggio's hitting streak in baseball, for example, raises serious questions about how the causes of relatively uncommon social events are typically studied. The hitting streak also directs our attention to considering models that generate a distribution of outcomes in a population of social units. The series of games leading to a football championship is relevant for approaching the role of complex chains of events in leading to a given outcome (where each individual sequence itself has a low probability of occurring). A model of changing templates suggests a more realistic way of thinking about the linkages over time from an initial starting point to a given outcome. Another sports example, based on the application of a binomial model to the outcome of a series of games, suggests that we should think of chance as a dependent variable that is influenced by—among other factors—the nature of the social structure. Also examined are the special problems in explaining any specific outcome.     

Local Control Strategy: Simple Analyses of Air Pollution Data Can Reveal Heterogeneity in Longevity Outcomes

Claims from observational studies that use traditional model specification searches often fail to replicate, partially because the available data tend to be biased. There is an urgent need for an alternative statistical analysis strategy, that is not only simple and easily understood but also is more likely to give reliable insights when the available data have not been designed and balanced. The alternative strategy known as local control first generates local, nonparametric effect‐size estimates (fair treatment comparisons) and only then asks whether the observed variation in these local estimates can be predicted from potential confounding factors. Here, we illustrate application of local control to a historical air pollution data set describing a “natural experiment” initiated by the federal Clean Air Act Amendments of 1970. Our reanalysis reveals subgroup heterogeneity in the effects of air quality regulation on elderly longevity (one size does not fit all), and we show that this heterogeneity is largely explained by socioeconomic and environmental confounders other than air quality.     

The influence of anthropogenic resources on multi-scale habitat selection by raccoons

With the continuing spread of urban areas, gaining a greater understanding of the effect of human presence on wildlife species is essential for wildlife managers. We determined the influence of anthropogenic resources on home range size and habitat selection of raccoons (Procyon lotor) during summer (June–August) 1996–2000 for 120 raccoons at three sites exposed to varying levels of urbanization and anthropogenic resources, specifically food. Home range estimates were larger (P < 0.05) at the rural site than the suburban and urban sites for both genders. We used compositional analysis to examine raccoon habitat selection at the second-order home range, second-order core area, and third-order home range scales. Woodland was consistently a highly-selected habitat type for both sexes at every spatial scale. Relative to other habitat types, habitat associated with human-related food (human use areas) was selected most often at the urban site, intermediately at the suburban site, and not selected at the rural site. Spatial scale also affected habitat selection. Human use areas were preferentially selected at the second- and third-order level at the urban site, third-order level only at the suburban site, and at neither level at the rural site. Additionally, intersexual differences in habitat selection were reduced at the urban site, with both sexes preferentially selecting for human use areas as well as woodland habitat. Smaller home ranges in urbanized environments are often attributed to the abundant and concentrated anthropogenic resources associated with human activity, but with little empirical support. Our habitat selection analyses followed our predictions that raccoon foraging is strongly influenced by the artificial distribution and abundance of human-related food. Male and female raccoons in urban areas reduce their foraging patterns and focus their foraging activity on anthropogenic foods.     

Online interval scheduling: randomized and multiprocessor cases

We consider the problem of scheduling a set of equal-length intervals arriving online, where each interval is associated with a weight and the objective is to maximize the total weight of completed intervals. An optimal 4-competitive algorithm has long been known in the deterministic case, but the randomized case remains open. We give the first randomized algorithm for this problem, achieving a competitive ratio of 3.5822. We also prove a randomized lower bound of 4/3, which is an improvement over the previous 5/4 result. Then we show that the techniques can be carried to the deterministic multiprocessor case, giving a 3.5822-competitive 2-processor algorithm, and a 4/3 lower bound for any number of processors. We also give a lower bound of 2 for the case of two processors. A preliminary version of this paper appeared in the Proceedings of COCOON 2007, LNCS, vol. 4598, pp. 176–186. The work described in this paper was fully supported by a grant from City University of Hong Kong (SRG 7001969), and NSFC Grant No. 70525004 and 70702030.     

Fast searching games on graphs

Given a graph, suppose that intruders hide on vertices or along edges of the graph. The fast searching problem is to find the minimum number of searchers required to capture all the intruders satisfying the constraint that every edge is traversed exactly once and searchers are not allowed to jump. In this paper, we prove lower bounds on the fast search number. We present a linear time algorithm to compute the fast search number of Halin graphs and their extensions. We present a quadratic time algorithm to compute the fast search number of cubic graphs.