Optimal Composite Markers for Time‐Dependent Receiver Operating Characteristic Curves with Censored Survival Data

Abstract. To increase the predictive abilities of several plasma biomarkers on the coronary artery disease (CAD)‐related vital statuses over time, our research interest mainly focuses on seeking combinations of these biomarkers with the highest time‐dependent receiver operating characteristic curves. An extended generalized linear model (EGLM) with time‐varying coefficients and an unknown bivariate link function is used to characterize the conditional distribution of time to CAD‐related death. Based on censored survival data, two non‐parametric procedures are proposed to estimate the optimal composite markers, linear predictors in the EGLM model. Estimation methods for the classification accuracies of the optimal composite markers are also proposed. In the article we establish theoretical results of the estimators and examine the corresponding finite‐sample properties through a series of simulations with different sample sizes, censoring rates and censoring mechanisms. Our optimization procedures and estimators are further shown to be useful through an application to a prospective cohort study of patients undergoing angiography.     

Till Death Do Us Part: Marital Status and U.S. Mortality Trends, 1986 – 2000

Although the association of being married and a lower mortality rate has been well established, most previous research on marital status and mortality did not consider potential change in this relationship over time. In this study, I adopted a survey cohort perspective to examine both overall and cause‐specific mortality trends by marital status from 1986 to 2000 in the United States. On the basis of data from the National Health Interview Survey‐Longitudinal Mortality Follow‐up (N = 517,314), I found that mortality generally decreased or remained stable for the married from the 1986 to 2000 NHIS cohort, except for diabetes deaths. There is evidence showing divergent mortality trends between the married and unmarried, especially the widowed. Race and gender variations are examined.     

Socioeconomic status and the dynamics of preferences for income inequality in the United States, 1978–2016

Although considerable evidence indicates that public preferences for income inequality and redistribution vary across socioeconomic groups (i.e., occupation and income), much less is known about the temporal dynamics of these preferences. The purpose of this study is (a) to examine whether the attitudinal distance between managerial/professional workers and unskilled manual workers has changed (converged or diverged) over time and to (b) explore the reasons for and implications of the dynamics of preferences in the past several decades. Using data from the General Social Survey 1978–2016 (23 time‐points; N = 27,211), this study finds that the influence of occupational class has lost some of its significance in shaping public preferences for income inequality and that the declining effect of occupation can be explained in part by the attitudinal convergence between better‐ and less‐educated citizens. Findings suggest that proequality coalitions across educational boundaries play a remedial role in bridging the occupational divide over government redistribution in the United States.     

Collective and Elective Ethnicity: Caste Among Urban Muslims in India

This article explores how the significance of ethnic identity can vary within a stable population, using caste among Muslims in Hyderabad, India as a case study. While some Hyderabadi Muslims are still embedded in ethnic networks, most now experience ethnicity as elective and do not rely on a corporate caste group for their social connections. This reflects a decline in the value of caste identities, which no longer provide economic or political resources. Increasingly, Muslims seek status through education, profession, or income. Thus, most Muslims in Hyderabad experience caste membership, identity, and networks in a weakened or attenuated way.     

社会资本及其本土化研究的几点思考

社会资本理论的提出,为各个学科提供了一种重要的解释范式,这正是它的重要价值所在。不过,社会资本理论也存在不少问题,如存在概念过窄、解释范围过宽之嫌;社会资本概念混乱;社会资本理论混淆了社会资本的来源与结果。国内在引入社会资本概念时,一定要从中国文化传统和当前社会转型的特殊性出发来研究社会资本,进而在此基础上建构出符合中国实际的社会资本理论。     

邓小平在中苏论战中的地位及贡献

客观地评价了邓小平在中苏论战中所处的历史地位,分析了邓小平在中苏论战中的突出贡献。指出在中苏论战中,邓小平坚决反对前苏联大党主义和大国主义,为维护民族尊严和国家主权做出了重大贡献,表现了中华民族不信邪、不怕压的大无畏精神。     

中国企业创建学习型组织的现状分析及对策研究——以江苏省连云港市企业为例

通过对连云港市企业创建学习型组织现状的实证调研,总结了连云港市企业在创建学习型组织过程中取得的一些经验,剖析了连云港市企业在创建学习型组织中存在的问题.最后,从普遍意义上提出了继续深化中国企业创建学习型组织的对策与建议.     

浙江省中小企业创新现状、问题及原因分析

中小企业占我国企业总数的99%以上,对GDP的贡献大约占六成,税收贡献约40%[1],同时解决大约75%就业[2].因此,中小企业发展问题至关重要.在当今转型升级背景下,如何提高中小企业创新能力,增强中小企业的竞争能力,成为国民经济发展战略的重要内容.本课题组立足浙江省中小企业创新现状和全面创新能力现状的调查,对中小企业创新过程中存在的问题进行深入分析,并对问题的深层次原因进行了探讨.     

存在荣誉效用的等级竞赛机制分析

针对存在荣誉效用的等级竞赛,研究在等级数已定的条件下,以总努力最大化为目标的竞赛设计者如何设置竞赛人数和各等级人数的问题。通过分析荣誉效用由于相邻等级容量调整所引起的变化对总努力的影响,得到结论：无论能力分布如何,将最高等级容量设置得很小,对竞赛设计者而言,是更合理的选择;当竞赛者的能力分布为凹函数时,将中低等级的容量设置得相对较大更合理;当能力分布为凸函数时,将中高等级的容量设置得相对较小更合理。     

Efficient estimation for the proportional hazards model with bivariate current status data

We consider efficient estimation of regression and association parameters jointly for bivariate current status data with the marginal proportional hazards model. Current status data occur in many fields including demographical studies and tumorigenicity experiments and several approaches have been proposed for regression analysis of univariate current status data. We discuss bivariate current status data and propose an efficient score estimation approach for the problem. In the approach, the copula model is used for joint survival function with the survival times assumed to follow the proportional hazards model marginally. Simulation studies are performed to evaluate the proposed estimates and suggest that the approach works well in practical situations. A real life data application is provided for illustration.