一种乘积估计量的研究

在抽样调查中恰当地使用辅助信息可以提高调查精度。当辅助变量x与调查指标y呈负相关关系时,乘积估计是构造总体均值估计传统的方法。本文提出了链式乘积估计量,并且得到了它的均方误差。在一定条件下,链式乘积估计量的效率高于传统估计量,并提供了一个数值例子。     

Bernstein算子线性组合加Jacobi权的收敛阶

通过引入一种新的范数给出了Ｂｅｒｎｓｔｅｉｎ算子线性组合加Ｊａｃｏｂｉ权的特征刻划。     

Functional calibration estimation by the maximum entropy on the mean principle

We extend the problem of obtaining an estimator for the finite population mean parameter incorporating complete auxiliary information through calibration estimation in survey sampling under a functional data framework. The functional calibration sampling weights of the estimator are obtained by matching the calibration estimation problem with the maximum entropy on the mean – MEM – principle. In particular, the calibration estimation is viewed as an infinite-dimensional linear inverse problem following the structure of the MEM approach. We give a precise theoretical setting and estimate the functional calibration weights assuming, as prior measures, the centred Gaussian and compound Poisson random measures. Additionally, through a simple simulation study, we show that the proposed functional calibration estimator improves its accuracy compared with the Horvitz–Thompson one.     

Estimating Mean Survival Time: When is it Possible?

For right‐censored survival data, it is well‐known that the mean survival time can be consistently estimated when the support of the censoring time contains the support of the survival time. In practice, however, this condition can be easily violated because the follow‐up of a study is usually within a finite window. In this article, we show that the mean survival time is still estimable from a linear model when the support of some covariate(s) with non‐zero coefficient(s) is unbounded regardless of the length of follow‐up. This implies that the mean survival time can be well estimated when the support of linear predictor is wide in practice. The theoretical finding is further verified for finite samples by simulation studies. Simulations also show that, when both models are correctly specified, the linear model yields reasonable mean square prediction errors and outperforms the Cox model, particularly with heavy censoring and short follow‐up time.     

有序响应变量的贝叶斯模型选择及其在COPD疾病防治中的应用

慢性阻塞性肺病(COPD)是一种发病率、死亡率都非常高的疾病,且COPD的诊断和严重程度分级依赖于肺功能的检查,但是由于肺功能检查仪器价格昂贵,使得这项检查在很多经济欠发达地区尤其是农村基层医院并没有普及。本文基于有序响应变量模型致力于研究一种便于基层和社区使用的可以初步判别COPD病情的模型,以期提高我国基层和社区的COPD 防治水平。利用贝叶斯变量选择方法和数据增强的潜变量策略得到了易于实施的Gibbs后验抽样算法。数值模拟分析进一步说明了本文提出的有序响应变量贝叶斯模型选择方法的有效性,实例分析得到了易于判别COPD严重程度的稀疏模型。     

大规模数据下基于充分降维的Leverage重要性抽样方法

随着信息技术的飞速发展,大规模数据在短时间内搜集并储存下来,为分析决策提供了巨大的信息量,也给统计建模带来了一定难度。对于样本容量大、变量个数少的数据,Leverage重要性抽样是一个简便可行的方法。本文发现,该方法中度量样本重要性的Leverage分数与因变量无关,而且在维度较大的情形下对样本没有区分程度,使得估计结果较差。为了同时考虑因变量和维度的影响,本文提出了基于充分降维的Leverage重要性抽样方法。该方法以不损失信息为前提,在充分降维的空间内重新计算Leverage分数,使得抽样更具有代表性。模拟数据分析显示,在样本容量较大的复杂数据中,相比于原始的Leverage重要性抽样方法,本文提出的方法可以降低估计的均方误差。三个实际数据也证实了该方法的可行性和有效性。     

Assessing an Adult Mental Health Clinic for Age-Related Issues

A community mental health clinic in an urban setting is the subject of an exploratory assessment to determine whether its 55 + patients are being treated according to the special diagnostic and treatment issues that tend to affect older people. Depression is found to be high on the list of diagnoses in this age group. As many as 40% of the group are considered to be at risk for suicide. The interrelationship of medical and psychological factors is emphasized. Sixty-two percent of those studied are determined to be socially isolated. Treatment techniques recommended in the literature for older patients are summarized, and it is determined that, while therapists at the clinic.have not been trained in geriatric psychology, many are instinctively utilizing these techniques. Recommendations are made for increased use of groupwork with older patients, inservice education for staff, attention to medical complications, improved outreach, and involvement of family.     

A simple in-host model for Mycobacterium tuberculosis that captures all infection outcomes

ABSTRACT

Tuberculosis infection can result in clearance, latent infection, or active disease, with slow or fast progression. A four-dimensional model of in-host tuberculosis infection includes macrophages, T lymphocytes, tuberculosis bacteria, and their interactions. Changes in the infection rate, cell-mediated immunity rate, macrophage loss rate, and bacteria killing rate most affect disease outcomes. Simulations show that a periodic solution can occur. When the infected macrophage killing rate is constant, a backward bifurcation exists and the system is globally stable.     

The sample size debate: response to Norman Blaikie

In his detailed response to our paper on sample size in qualitative research, Norman Blaikie raises important issues concerning conceptual definitions and taxonomy. In particular, he points out the problems associated with a loose, generic application of adjectives such as ‘qualitative’ or ‘inductive’. We endorse this concern, though we suggest that in some specific contexts a broad categorization may be more appropriate than a more nuanced distinction – provided that it is clear in which sense the terms are employed. However, other concepts, such as saturation, do not lend themselves to generic use, and require a more detailed conceptualization. Blaikie’s analysis also makes it clear that meaningful discussion of sample size in qualitative research cannot occur with reference to an undifferentiated conception of the nature of qualitative research; clear distinctions need to be made within this approach in terms of methodology, ontological and epistemological assumptions and broader research paradigms.     

A generalization of the logistic linear'model

Consider the logistic linear model, with some explanatory variables overlooked. Those explanatory variables may be quantitative or qualitative. In either case, the resulting true response variable is not a binomial or a beta-binomial but a sum of binomials. Hence, standard computer packages for logistic regression can be inappropriate even if an overdispersion factor is incorporated. Therefore, a discrete exponential family assumption is considered to broaden the class of sampling models. Likelihood and Bayesian analyses are discussed. Bayesian computation techniques such as Laplacian approximations and Markov chain simulations are used to compute posterior densities and moments. Approximate conditional distributions are derived and are shown to be accurate. The Markov chain simulations are performed effectively to calculate posterior moments by using the approximate conditional distributions. The methodology is applied to Keeler's hardness of winter wheat data for checking binomial assumptions and to Matsumura's Accounting exams data for detailed likelihood and Bayesian analyses.