Empirical likelihood weighted composite quantile regression with partially missing covariates

This paper develops a novel weighted composite quantile regression (CQR) method for estimation of a linear model when some covariates are missing at random and the probability for missingness mechanism can be modelled parametrically. By incorporating the unbiased estimating equations of incomplete data into empirical likelihood (EL), we obtain the EL-based weights, and then re-adjust the inverse probability weighted CQR for estimating the vector of regression coefficients. Theoretical results show that the proposed method can achieve semiparametric efficiency if the selection probability function is correctly specified, therefore the EL weighted CQR is more efficient than the inverse probability weighted CQR. Besides, our algorithm is computationally simple and easy to implement. Simulation studies are conducted to examine the finite sample performance of the proposed procedures. Finally, we apply the new method to analyse the US news College data.     

Accounting for clinical covariates and interactions in ranking genomic markers using ROC

In biomedical research, profiling is now commonly conducted, generating high-dimensional genomic measurements (without loss of generality, say genes). An important analysis objective is to rank genes according to their marginal associations with a disease outcome/phenotype. Clinical-covariates, including for example clinical risk factors and environmental exposures, usually exist and need to be properly accounted for. In this study, we propose conducting marginal ranking of genes using a receiver operating characteristic (ROC) based method. This method can accommodate categorical, censored survival, and continuous outcome variables in a very similar manner. Unlike logistic-model-based methods, it does not make very specific assumptions on model, making it robust. In ranking genes, we account for both the main effects of clinical-covariates and their interactions with genes, and develop multiple diagnostic accuracy improvement measurements. Using simulation studies, we show that the proposed method is effective in that genes associated with or gene–covariate interactions associated with the outcome receive high rankings. In data analysis, we observe some differences between the rankings using the proposed method and the logistic-model-based method.     

Meta-analysis of Proportions of Rare Events–A Comparison of Exact Likelihood Methods with Robust Variance Estimation

The conventional random effects model for meta-analysis of proportions approximates within-study variation using a normal distribution. Due to potential approximation bias, particularly for the estimation of rare events such as some adverse drug reactions, the conventional method is considered inferior to the exact methods based on binomial distributions. In this article, we compare two existing exact approaches—beta binomial (B-B) and normal-binomial (N-B)—through an extensive simulation study with focus on the case of rare events that are commonly encountered in medical research. In addition, we implement the empirical (“sandwich”) estimator of variance into the two models to improve the robustness of the statistical inferences. To our knowledge, it is the first such application of sandwich estimator of variance to meta-analysis of proportions. The simulation study shows that the B-B approach tends to have substantially smaller bias and mean squared error than N-B for rare events with occurrences under 5%, while N-B outperforms B-B for relatively common events. Use of the sandwich estimator of variance improves the precision of estimation for both models. We illustrate the two approaches by applying them to two published meta-analysis from the fields of orthopedic surgery and prevention of adverse drug reactions.     

\lambda -numbers of several classes of snarks

In the paper we study \(\lambda \) -numbers of several classes of snarks. We show that the \(\lambda \) -number of each Blanu \(\breve{s}\) a snark, Flower snark and Goldberg snark is \(6\) . For \(n\ge 2\) , we show that there is a dot product of \(n\) Petersen graphs such that its \(\lambda \) -number is 6.     

战略联盟伙伴间特征与联盟管理实践转移

为了探究联盟管理方法、流程和技能如何在联盟伙伴之间转移,引入联盟管理实践,并对其进行界定。基于知识基础理论,探讨战略联盟伙伴间的行为要素（共同行动）、文化要素（文化相似性）和关系要素（关系团结）3个特征对联盟管理实践获取的影响,研究联盟管理实践转移的3个阶段（即联盟管理实践获取、联盟管理实践应用和联盟管理实践内化）之间的关系。通过对205家联盟企业进行问卷调研获得研究样本,每个企业包括2个关键信息提供者,运用逐步层级回归对假设进行检验。研究结果表明,共同行动和文化相似性均会促进联盟管理实践获取,关系团结与联盟管理实践获取的倒U形相关关系不显著,联盟管理实践应用在联盟管理实践获取与联盟管理实践内化之间起部分中介作用。     

出生人口性别比失调与从严控制人口中的误导与失误

中国人口控制走过的弯路与21年的出生人口性别比失调,暴露出人口研究与决策中的一些问题。文章分析了这些问题的产生原因及其造成的后果,试图澄清在人口控制与出生性别比问题上的"是是非非"。     

劳动力流动:中国农村收入增长的新因素

劳动力流出究竟会促进还是削弱当地的发展是 1 990年代国内外学术界争论的一个焦点。本文运用中国 2 0 0 0年普查数据以及分县的时间序列数据 ,考察了人口流动趋势及其对当地劳动力和农村收入增长的影响。结果显示 ,1 990年代的大量劳动力流出对粮食生产影响并不显著 ,输出地的肉类产品反而有了较快的增长 ,劳动力流动已成为不发达地区农村收入新的增长点 ,正逐步发挥其消除贫困制约地区差距的效应。     

A study of urban population migration and urbanization in China]

Under the auspices of the Chinese Academy of Social Sciences and UN funding, a study was undertaken to assess Chinese urban migration and urbanization. A 2% random sample was taken of 74 cities of varying densities (divided into 5 categories ranging from "especially large city" to "town") from 16 provinces. This encompassed 23,895 households, 1,643 collectives and 100,167 people. Major data include: 38% of the subjects had migrated at least once; 7.58% lived away from home for at least 1 year; 23.98% were temporarily away from home at the time of the study; 3.6% were at home. With the exception of the "especially large city," which absorbed 46.5% of urban migrants, more migrants entered "towns" than they did "large city." Migration to the "especially large city" fell from 56.6% in the 1950s to 32.5% between 1981-86, whereas migration into the other categories increased. For example, population movement into towns jumped from 12.3% in the 1950s to 28.6% in 1981-86. In all 5 categories, intra-province migration was larger than inter-province migration. Over half of the urban migrants moved from villages to towns. More men migrated to cities than women, but slightly more women than men migrated from villages to cities (due to marriage customs). 56.6% of migrants were between 15-30 years; 23.34% were workers; 21.54% were farmers. Reasons given for moving were many, but the most often cited was work related. Work related moves often meant that such migration was dictated, rather than voluntary. Also, social, economic and political upheavals directly affected the pattern of urban migration from 1949-86. Current government policy is to develop smaller cities and to limit the growth of already densely populated areas. Until cities can provide adequate housing, food and jobs for its inhabitants, governmental intervention in some areas will continue to be necessary.     

Comparative analysis of the index of our population life quality]

A comprehensive method of calculating and measuring a country's or an area's health and literacy levels is examined. The method, known as population quality life inference (PQLI), was used to determine which of China's provinces has the highest and the lowest degree of population quality. The PQLI indicates infant mortality, average life expectancy of 1 year olds, and literacy rates of those 15 years and older. Because developing countries traditionally have high rates of infant mortality and illiteracy and low life expectancy rates during their industrialization, measuring the degree of population quality of life improvement of such countries during this period was found to be significant. These factors (infant mortality, illiteracy, and life expectancy) will improve substantially as industrialization continues. In order to compare various areas, these 3 factors must be changed into "inferences" 0-100, "0" representing the lowest population quality and "100" the highest. These 3 inferences must then be averaged in order to calculate the PQLI. For example: life expectancy value 77 (highest in the world) minus 38 (lowest)/100 = .39. In order to measure the value of India's life expectancy: value of 1-year-old's life expectancy = 56 (1-year-old's life expectancy in India) minus 38/.39 = 46. The value of adult illiteracy does not need to be changed. Thus, the actual comparison will be based on the values of the 3 inferences. Using this method of calculation, it is concluded that the PQLI analysis indicated that Peking (93.04) is the highest in China and Yumnan Province (60.72) is the lowest.     

统筹养老金双基数征缴方案设计

双基数征缴的目的在于通过调整企业间的养老负担,保证统筹养老金的征缴率和征缴量。当前的统筹养老金可以分解为基础养老金和转轨成本两部分。在双基数征缴的具体实施过程中,可以以职工工资总额的7%-8%征缴基础养老金, 以企业税前利润的1%征缴转轨成本部分。以利润为基数征缴的部分是政府向企业的强制性借款,政府在日后应有计划地逐步偿还。