Asa Advisory Committees to Federal Statistical Agencies: A Report of the Committee on Committees of the American Statistical Association

(Editor's Note: The following report on ASA advisory committees was prepared by Edwin D. Goldfield as part of a continuing study of the Association's committees and Sections being conducted by the Committee on Committees. Because of the importance of advisory committees to Federal agencies, the report is included here for the information of the members of the Association.)     

A Derivation of a Useful Consequence of the Gauss-Markov Theorem

The Gauss-Markov (GM) theorem and its many consequences are well known. It has been our experience that graduate students in linear model courses benefit substantially by deriving these consequences, all of which form natural corollaries to the GM theorem. In deriving any one lemma, the student is compelled to reuse many other GM-based lemmas. In this article we present another example of such a lemma that arises in the forward-elimination variable selection procedure. The derivation given is considered to be pedagogically useful.     

Report of the ASA Technical Panel on the Census Undercount

A new model is proposed for the joint distribution of paired survival times generated from clinical trials and certain reliability settings. The new model can be considered an extension to the bivariate exponential models studied in the literature. Here, a more flexible bivariate Weibull model will be derived, and two exact parametric tests for testing the equality of marginal survival distributions are developed.     

The determinacy of the regression factor score predictor based on continuous parameter estimates from categorical variables

The estimation of population parameters of the continuous common factor model from categorical observed variables is meanwhile regularly performed. It is shown that the formula for the calculation of the determinacy of the regression factor score predictor from the estimated model parameters has to be adapted under these conditions. A method for the calculation of this determinacy from the model parameters of the continuous population factor model based on categorical variables is proposed and evaluated by means of simulated population data. It turns out that using the uncorrected formula can lead to serious overestimation of determinacy for categorical variables.     

Exact and Monte Carlo calculations of integrated likelihoods for the latent class model

The latent class model or multivariate multinomial mixture is a powerful approach for clustering categorical data. It uses a conditional independence assumption given the latent class to which a statistical unit is belonging. In this paper, we exploit the fact that a fully Bayesian analysis with Jeffreys non-informative prior distributions does not involve technical difficulty to propose an exact expression of the integrated complete-data likelihood, which is known as being a meaningful model selection criterion in a clustering perspective. Similarly, a Monte Carlo approximation of the integrated observed-data likelihood can be obtained in two steps: an exact integration over the parameters is followed by an approximation of the sum over all possible partitions through an importance sampling strategy. Then, the exact and the approximate criteria experimentally compete, respectively, with their standard asymptotic BIC approximations for choosing the number of mixture components. Numerical experiments on simulated data and a biological example highlight that asymptotic criteria are usually dramatically more conservative than the non-asymptotic presented criteria, not only for moderate sample sizes as expected but also for quite large sample sizes. This research highlights that asymptotic standard criteria could often fail to select some interesting structures present in the data.     

统计显著性:一个被误读的P值——基于美国统计学会的声明

美国统计学会关于统计显著性与P值的官方声明发布之后,再次引发国内外研究学者对P值的广泛关注。在介绍国内统计教材中假设检验的基本内容和步骤的基础上,以硬币投掷与背影识人为例直观性解释P值、统计显著性与统计功效等相关概念,并引用心理学统计经典调查案例分析P值被误读的原因。同时,基于美国统计学会的声明,给出正确使用P值的建议。     

U.S. Institutional Representation on Editorial Boards of U.S. Statistics Journals

This article presents two different, but similar, rankings of U.S. statistics faculties according to their current representation on editorial boards of 14 U.S. statistics journals. The data show that there is a widespread balance of individuals, institutions, and geographical locations in the U.S. with respect to publication officers.     

对增加农民收入的思索

近几年农民收入增长缓慢，有的农产品增产不增收，农民这一弱势群体已引起上下一致关注，作为农村经济调查者，也该为之思索一番。 农民的含义：狭义指从事种植业的劳动人员，广义指从事农、林、牧、渔业的劳动人员，现在统计上简称的“农民收入”，实际是指“农村居民”的人均收入，既包括农村非劳动力，又包括已向其他行业转移的劳动力(户籍尚在农村)。本文所提农民，主要指狭义上的农民。 农村大背景：集体土地实行家庭联产承包制度，其经营期限再延长30年，这是农村政策的核心内容，是考虑农村问题的大前提；农业是国民经济的基础…     

产业关联效应测度理论辨析

基于投入产出模型可以构建多种产业关联效应测度方法,但不同方法间的区别与联系尚未获得足够重视和透彻把握。本文探讨了常规不加权产业关联测度的真实内涵和局限性,并由经济分析入手探讨和论证了相应的加权测度形式;将简单和加权的两种方式归纳为产业关联的相对（边际或平均）测度和绝对（规模）测度,着重辨析其经济内涵的异同点;进而分别采用Leontief和Ghosh模型体系,从需求拉动和供给推动两个角度测度后向和前向产业关联效应。结合我国投入产出数据的实证分析表明,不同测度方法给出的结果均有其经济分析价值;但相对而言,考虑规模因素的加权测度方法在刻画实际的产业关联效应方面显得更为客观、可信。     

对基本单位名录库建设的思考

难以完整的现状 近几年来，为加强行政监管，工商、税务、技监、统计、民政、编制等部门均从部门的职能出发，依据自身的行政记录，建立了各自的单位名录库。但由于多方面的原因，部门间的名录库没有相互衔接和信息共享，基本单位名录库难以适时更新，从而大大降低了其应有的作用。掌握各类基本单位的数量以及各类单位的组织形式、经济成分构成、规模结构、生产要素配置和地区、行业分布等情况，仍需依赖五年一次的基本单位普查，这不仅浪费人力和财力，也增加了基层负担。 从第二次基本单位普查过程中基普库与几个部门库的核对情况看…     

Nontraditional Approaches to Assessing Journal Importance: Case Study of Russian Journals on Earth Sciences

Abstract

Currently, a broad set of bibliometric indices for evaluation of journal quality, the level of its internationalization, and interactions between serials are available. However, some cases require one to obtain more precise methods to assess journals, e.g., when estimating new journals, selecting them into international citation databases, increasing their rank, broadening their readership, or optimizing editorial policy. To cope with new challenges, new approaches have been introduced including altmetrics and various analyses of editorial board composition and impact. Analyzing altmetrics and studying geographic diversity on editorial boards, comparing these data with that of journal authors, and evaluating research efficiency of editorial board members enables one to use new approaches to assess the rank of serials. This paper considers current research reports on altmetrics, geographic diversity, and publication activity of editorial board members of top journals on Earth Sciences published in the Russian Federation and recognized as international serials. It was found that such altmetrics as abstract views, full text views, and a number of viewed papers are now available as additional metrics to evaluate journal quality. Furthermore, provided that the data on altmetrics are complete, they can be considered as a specific predictor for future citations and other citation-based indices. Editorial board analyses demonstrate a correlation of bibliometric indices of editorial board members with that of a journal. As for geographic diversity, only half of the analyzed journals demonstrate a strong correlation between the number of foreign members on editorial boards and geographic representativeness of a journal, mainly due to the local nature of geosciences and trade secret mode in much of mining research. Generally, the obtained findings highlighted the efficiency of new approaches to evaluate journal quality both in the case of the absence of citation-based journal metrics and as an additional tool to traditional indices.     

完善工业发展速度新计算方法的建议

工业发展速度采用价格指数缩减法，是适应新形势要求的统计方法重大改革，必须在统计实践中认真落实执行，但新方法也有值得完善的地方。 1.不可比产品的核算问题。 基期没有生产而报告期新生产的产品为不可比产品，不可比产品是不可避免的，报告期工业增加值应是可比产品工业增加值和不可比产品工业增加值之和，报告期必须统计不可比产品的工业总产值和增加值进而计算工业发展速度。不可比产品工业增加值不存在价格变动的影响，用价格指数缩减报告期全部工业增加值，就把不可比产品当成了可比产品，当价格指数大于100%，不可比产品的…     

Preparing Statisticians for Careers in Industry: Report of the ASA Section on Statistical Education Committee on Training of Statisticians for Industry

This report provides guidelines for universities to consider in developing programs for training statisticians who will work in industry. Useful information for students who are considering industrial employment is also included. The recommended programs focus on real problems and the statistical theory and methodology that are useful in their solution. Technical competence is only one of many factors that industry considers when hiring and promoting statisticians. When a statistician leaves school, his or her skills and experiences should include statistical knowledge, practical problem solving, consulting practice, and the ability to communicate orally and in writing with nonstatisticians. There are many details that must be worked out (e.g., content of specific courses and organization of consulting internship programs), and it is hoped that the statistical societies and universities will form committees, hold conferences, and develop programs to address these issues further. Many of our recommendations apply more broadly to the training of all types of practicing statisticians.     

An empirical bayes approach to estimating the probability of correct selection

In the problem of selecting the best of k populations, Olkin, Sobel, and Tong (1976) have introduced the idea of estimating the probability of correct selection. In an attempt to improve on their estimator we consider anempirical Bayes approach. We compare the two estimators via analytic results and a simulation study.     

做好企业统计工作 为企业管理服务

社会上有这样一种误传:统计,就是七分统计三分估计。我认为应该改为十分统计,十二分的估计。所谓十分统计,就是用百分之百准确、真实的数据进行科学地汇总和深加工,来反映企业生产经营活动过程中的发展状     

物价对消费行为影响的实证研究

文章利用1985~2003年我国城市居民人均实际可支配收入和人均消费支出,结合西方消费理论,通过建立经济计量模型,揭示了物价对城市居民不同收入群体消费行为的影响,并指出其经济意义。     

中国出口贸易中的服务含量分析

本文基于OECD-WTO的跨国投入-产出表（ICIO）与TiVA数据,采用增加值贸易方法,测算出中国贸易出口中的服务含量所占比重大约是传统BOP统计口径的两倍,其中来自国外的服务含量占比趋于上升、来自国内的服务含量占比趋于下降;在双边贸易中,中国的增加值贸易出口中来自美国、日本、德国、韩国等发达经济体的服务含量较高。本文的启示是,在全球产业链与价值链分工深化背景下,中国国内服务对增加值贸易的贡献偏低并趋于下降,可能原因在于国内服务业国际竞争力不强,因此,政策的着力点应是以促进竞争为导向深化国内服务领域改革,促进国内产业链和价值链环节从低端向高端演进。     

Bayesian Approaches to Nonparametric Estimation of Densities on the Unit Interval

This paper investigates nonparametric estimation of density on [0, 1]. The kernel estimator of density on [0, 1] has been found to be sensitive to both bandwidth and kernel. This paper proposes a unified Bayesian framework for choosing both the bandwidth and kernel function. In a simulation study, the Bayesian bandwidth estimator performed better than others, and kernel estimators were sensitive to the choice of the kernel and the shapes of the population densities on [0, 1]. The simulation and empirical results demonstrate that the methods proposed in this paper can improve the way the probability densities on [0, 1] are presently estimated.