Estimation of cumulative incidence functions when the lifetime distributions are uniformly stochastically ordered

In the competing risks problem an important role is played by the cumulative incidence function (CIF), whose value at time t is the probability of failure by time t from a particular type of risk in the presence of other risks. Assume that the lifetime distributions of two populations are uniformly stochastically ordered. Since this ordering may not hold for the empiricals due to sampling variability, it is natural to estimate these distributions under this constraint. This will in turn affect the estimation of the CIFs. This article considers this estimation problem. We do not assume that the risk sets in the two populations are related, give consistent estimators of all the CIFs and study the weak convergence of the resulting processes. We also report the results of a simulation study that show that our restricted estimators outperform the unrestricted ones in terms of mean square error. A real life example is used to illustrate our theoretical results.     

Some remarks on structural inference applied to weibull distribution

The structural probability distribution of the parameters of the two-parameter Weibull distribution is derived directly from considerations of the group structure of its density function. In the process we compare the structural method of inference with the confidence interval approach and reveal their similarities and differences. Structural prediction densities of arbitrary ordered statistics from Weibull distributions are also given to complement a previous work by Bury and Burnholtz.     

Arab immigrants in the United States: how and why do returns to education vary by country of origin?

Using U.S. census data, the authors analyze the earnings of Arab males who completed their schooling before migrating to the United States. There is little return to precollege education, but education beyond 12 years is rewarded highly. Although Arabs share a common ethnicity, they are not a homogeneous group. Returns to education vary significantly by source-country, e.g., high for immigrants from Kuwait, low for Yemeni immigrants. Returns are related to economic development in the source-country and to pupil/teacher ratios. These findings have implications for immigration policy and point to the hazards of generalizing on the basis of ethnicity.     

Marital status,sex, and suicide: new longitudinal findings and Durkheim’s marital status propositions*

Abstract

The purpose of the paper was to use new longitudinal data to investigate the impact of marital status on suicide, and test Durkheim’s marital status propositions. Previous research found marital status was associated with suicide, but some studies neglected sex, most of the research was cross-sectional, and divorce and separated statuses were often combined. Data were obtained from the latest release of the U.S. National Longitudinal Mortality Study (NLMS), consisting of the largest U.S. sample of suicide victims, 2,013 out of 1.5 million persons. Proportional hazards and logistic regression models were fitted to the NLMS data based on follow-up from 1990 to 2011. Results showed that when combined, the divorced and separated were over 88% more likely to suicide than the married (ARR = 1.886, CI = 1.649, 2.156). When split, the divorced had suicide risk that was over 97% higher than that of the married (ARR = 1.973, CI = 1.711, 2.274). Separated individuals experienced suicide risk that was nearly 52% greater than that of the married (ARR = 1.515, CI = 1.130, 2.037). The Mountain and southern census divisions had higher suicide risks than New England. Discussion focused mainly on Durkheim’s theory of suicide.     

Socioenvironmental Threats to Pastoral Livelihoods: Risk Perceptions in the Altay and Tianshan Mountains of Xinjiang,China

Subjective risk perceptions give rise to unique policy implications as they reflect both the expectation of risk exposure and the ability to mitigate or cope with the adverse impacts. Based on data collected from semistructured interviews and iterative ranking exercises with 159 households in the Altay and Tianshan Mountains of Xinjiang, China, this study investigates and explains the risks with respect to a seriously understudied population and location. Using both geostatistical and econometric methods, we show that although fear of environmental crisis is prevalent among our respondents, recently implemented pastoral conservation, sedentarization, and development projects are more likely to be ranked as the top concerns among affected households. In order to reduce these concerns, future pastoral policy must be built on the livestock economy, and intervention priority should be given to the geographic areas identified as risk hot spots. In cases where pastoralists have to give up their pastures, the transition to other comparable livelihood strategies must be enabled by creating new opportunities and training pastoralists to acquire the needed skills.     

On quantile processes for m-dependent Rv's

We obtain strong and weak approximations for quantile processes for m-dependent random variables. We apply our results in the two-sample to obtain condidence bands for quantile plots and prove some CHERNOFF-SAVAGE theorems for m-dependent random variables     

Nonparametric Tests for Comparing Several Coefficients of Variation

We consider the problem of comparing (k + 1) coefficients of variation. We are interested in testing the null hypothesis that the coefficients of variation are equal against each of the alternatives: (a) some populations have different coefficients of variation and (b) the coefficients of variation are ordered. Three nonparametric test statistics are proposed and their asymptotic theory is developed. We compared the proposed tests together with another parametric test using two Monte Carlo studies to estimate their probabilities of Type I error and powers. An illustration of the proposed tests using a real data set is given.     

The evolution and problems of model management research

Models are a key resource for organizational decision making. The diversity, complexity, and reusability of this resource result in the need for model management systems. The construction of a model management system involves a modeling task dimension and a design level dimension. The modeling task dimension consists of model formulation, model representation, and model processing. The design level dimension addresses the architectural requirements of a system from a user's standpoint and a computer system's standpoint.The numerous architectures suggested in the model management systems literature address isolated areas identified by these dimensions. The research surveyed in this paper indicates that the primary focus has been on the system view of model representation. Before model management systems can be widely used in organizations, model management researchers must explore systems that address all areas of the task and design level dimensions.In this paper, we identify and justify the necessary dimensions of model management research. Next, the existing model management research is critically reviewed. Finally, neglected research areas are discussed, and investigations necessary for the development of integrated model management systems are suggested.     

Some limit theorems for uniform and exponential spacings

The asymptotic distributions of spacing statistics based on exponential and uniform spacings are studied, and the corresponding rates of convergence are also given.