On pseudo-values for regression analysis in competing risks models

For regression on state and transition probabilities in multi-state models Andersen et al. (Biometrika 90:15–27, 2003) propose a technique based on jackknife pseudo-values. In this article we analyze the pseudo-values suggested for competing risks models and prove some conjectures regarding their asymptotics (Klein and Andersen, Biometrics 61:223–229, 2005). The key is a second order von Mises expansion of the Aalen-Johansen estimator which yields an appropriate representation of the pseudo-values. The method is illustrated with data from a clinical study on total joint replacement. In the application we consider for comparison the estimates obtained with the Fine and Gray approach (J Am Stat Assoc 94:496–509, 1999) and also time-dependent solutions of pseudo-value regression equations.     

A Bayesian approach to non-parametric monotone function estimation

Summary.  The paper proposes two Bayesian approaches to non-parametric monotone function estimation. The first approach uses a hierarchical Bayes framework and a characterization of smooth monotone functions given by Ramsay that allows unconstrained estimation. The second approach uses a Bayesian regression spline model of Smith and Kohn with a mixture distribution of constrained normal distributions as the prior for the regression coefficients to ensure the monotonicity of the resulting function estimate. The small sample properties of the two function estimators across a range of functions are provided via simulation and compared with existing methods. Asymptotic results are also given that show that Bayesian methods provide consistent function estimators for a large class of smooth functions. An example is provided involving economic demand functions that illustrates the application of the constrained regression spline estimator in the context of a multiple-regression model where two functions are constrained to be monotone.     

Measuring people's trust

Summary. We measure trust and trustworthiness in British society with a newly designed experiment using real monetary rewards and a sample of the British population. The study also asks the typical survey question that aims to measure trust, showing that it does not predict 'trust' as measured in the experiment. Overall, about 40% of people were willing to trust a stranger in our experiment, and their trust was rewarded half of the time. Analysis of variation in the trust behaviour in our survey suggests that trusting is more likely if people are older, their financial situation is either 'comfortable' or 'difficult' compared with 'doing alright' or 'just getting by', they are a homeowner or they are divorced, separated or never married compared with those who are married or cohabiting. Trustworthiness also is more likely among subjects who are divorced or separated relative to those who are married or cohabiting, and less likely among subjects who perceive their financial situation as 'just getting by' or 'difficult'. We also analyse the effect of attitudes towards risks on trust.     

Parametric non-mixture cure models for schedule finding of therapeutic agents

We propose a phase I clinical trial design that seeks to determine the cumulative safety of a series of administrations of a fixed dose of an investigational agent. In contrast with traditional phase I trials that are designed solely to find the maximum tolerated dose of the agent, our design instead identifies a maximum tolerated schedule that includes a maximum tolerated dose as well as a vector of recommended administration times. Our model is based on a non-mixture cure model that constrains the probability of dose limiting toxicity for all patients to increase monotonically with both dose and the number of administrations received. We assume a specific parametric hazard function for each administration and compute the total hazard of dose limiting toxicity for a schedule as a sum of individual administration hazards. Throughout a variety of settings motivated by an actual study in allogeneic bone marrow transplant recipients, we demonstrate that our approach has excellent operating characteristics and performs as well as the only other currently published design for schedule finding studies. We also present arguments for the preference of our non-mixture cure model over the existing model.     

Peer group ties and executive compensation networks

Publicly traded firms in the US typically determine C.E.O. compensation by benchmarking the pay of their C.E.O.s against the pay of C.E.O.s in “peer” firms. Consequently, executive compensation is influenced not only by firm-level characteristics, but also by the selection and actions of the firm’s immediate peers as well as by the structure of the executive compensation network overall. Analyzing compensation peer group choices made by the same 1183 firms for F.Y. 2007, 2008 and 2009, we find that while the typical compensation peer is similar in size and industry to the firm that chose it, deviations from this norm are common, especially among larger firms, and tend to be towards larger firms with better paid CEOs. Further analysis shows that firms who pay CEOs well relative to the pay that would be predicted from their revenues, return on assets, and industry tend to have greater aspiration bias in their group of named peers.     

Trends in gender segregation in the choice of science and engineering majors

Numerous theories have been put forward for the high and continuing levels of gender segregation in science, technology, engineering, and mathematics (STEM) fields, but research has not systematically examined the extent to which these theories for the gender gap are consistent with actual trends. Using both administrative data and four separate longitudinal studies sponsored by the U.S. Department of Education’s National Center for Education Statistics (NCES), we evaluate several prominent explanations for the persisting gender gap in STEM fields related to mathematics performance and background and general life goals, and find that none of them are empirically satisfactory. Instead, we suggest that the structure of majors and their linkages to professional training and careers may combine with gender differences in educational goals to influence the persisting gender gap in STEM fields. An analysis of gendered career aspirations, course-taking patterns, and pathways to medical and law school supports this explanation.     

Social and behavioral skills and the gender gap in early educational achievement

Though many studies have suggested that social and behavioral skills play a central role in gender stratification processes, we know little about the extent to which these skills affect gender gaps in academic achievement. Analyzing data from the Early Child Longitudinal Study-Kindergarten Cohort, we demonstrate that social and behavioral skills have substantively important effects on academic outcomes from kindergarten through fifth grade. Gender differences in the acquisition of these skills, moreover, explain a considerable fraction of the gender gap in academic outcomes during early elementary school. Boys get roughly the same academic return to social and behavioral skills as their female peers, but girls begin school with more advanced social and behavioral skills and their skill advantage grows over time. While part of the effect may reflect an evaluation process that rewards students who better conform to school norms, our results imply that the acquisition of social and behavioral skills enhances learning as well. Our results call for a reconsideration of the family and school-level processes that produce gender gaps in social and behavioral skills and the advantages they confer for academic and later success.     

An Integrated Risk Assessment Model of Township‐Scaled Land Subsidence Based on an Evidential Reasoning Algorithm and Fuzzy Set Theory

Land subsidence risk assessment (LSRA) is a multi‐attribute decision analysis (MADA) problem and is often characterized by both quantitative and qualitative attributes with various types of uncertainty. Therefore, the problem needs to be modeled and analyzed using methods that can handle uncertainty. In this article, we propose an integrated assessment model based on the evidential reasoning (ER) algorithm and fuzzy set theory. The assessment model is structured as a hierarchical framework that regards land subsidence risk as a composite of two key factors: hazard and vulnerability. These factors can be described by a set of basic indicators defined by assessment grades with attributes for transforming both numerical data and subjective judgments into a belief structure. The factor‐level attributes of hazard and vulnerability are combined using the ER algorithm, which is based on the information from a belief structure calculated by the Dempster‐Shafer (D‐S) theory, and a distributed fuzzy belief structure calculated by fuzzy set theory. The results from the combined algorithms yield distributed assessment grade matrices. The application of the model to the Xixi‐Chengnan area, China, illustrates its usefulness and validity for LSRA. The model utilizes a combination of all types of evidence, including all assessment information—quantitative or qualitative, complete or incomplete, and precise or imprecise—to provide assessment grades that define risk assessment on the basis of hazard and vulnerability. The results will enable risk managers to apply different risk prevention measures and mitigation planning based on the calculated risk states.     

Voting Technology,Political Responsiveness,and Infant Health: Evidence From Brazil

This paper studies the introduction of electronic voting technology in Brazilian elections. Estimates exploiting a regression discontinuity design indicate that electronic voting reduced residual (error‐ridden and uncounted) votes and promoted a large de facto enfranchisement of mainly less educated citizens. Estimates exploiting the unique pattern of the technology's phase‐in across states over time suggest that, as predicted by political economy models, it shifted government spending toward health care, which is particularly beneficial to the poor. Positive effects on both the utilization of health services (prenatal visits) and newborn health (low‐weight births) are also found for less educated mothers, but not for the more educated.     

Boosting multi-state models

One important goal in multi-state modelling is to explore information about conditional transition-type-specific hazard rate functions by estimating influencing effects of explanatory variables. This may be performed using single transition-type-specific models if these covariate effects are assumed to be different across transition-types. To investigate whether this assumption holds or whether one of the effects is equal across several transition-types (cross-transition-type effect), a combined model has to be applied, for instance with the use of a stratified partial likelihood formulation. Here, prior knowledge about the underlying covariate effect mechanisms is often sparse, especially about ineffectivenesses of transition-type-specific or cross-transition-type effects. As a consequence, data-driven variable selection is an important task: a large number of estimable effects has to be taken into account if joint modelling of all transition-types is performed. A related but subsequent task is model choice: is an effect satisfactory estimated assuming linearity, or is the true underlying nature strongly deviating from linearity? This article introduces component-wise Functional Gradient Descent Boosting (short boosting) for multi-state models, an approach performing unsupervised variable selection and model choice simultaneously within a single estimation run. We demonstrate that features and advantages in the application of boosting introduced and illustrated in classical regression scenarios remain present in the transfer to multi-state models. As a consequence, boosting provides an effective means to answer questions about ineffectiveness and non-linearity of single transition-type-specific or cross-transition-type effects.