Sparse group lasso for multiclass functional logistic regression models

Sparsity-inducing penalties are useful tools for variable selection and are also effective for regression problems where the data are functions. We consider the problem of selecting not only variables but also decision boundaries in multiclass logistic regression models for functional data, using sparse regularization. The parameters of the functional logistic regression model are estimated in the framework of the penalized likelihood method with the sparse group lasso-type penalty, and then tuning parameters for the model are selected using the model selection criterion. The effectiveness of the proposed method is investigated through simulation studies and the analysis of a gene expression data set.     

An m-estimation-based model selection criterion with a data-oriented penalty

In Wu and Zen (1999), a linear model selection procedure based on M-estimation is proposed, which includes many classical model selection criteria as its special cases, and it is shown that the selection procedure is strongly consistent for a variety of penalty functions. In this paper, we will investigate its small sample performances for some choices of fixed penalty functions. It can be seen that the performance varies with the choice of the penalty. Hence, a randomized penalty based on observed data is proposed, which preserves the consistency property and provides improved performance over a fixed choice of penalty functions.     

1906-1908年的郑孝胥与预备立宪公会

考虑到我国未来年度的OD分布预测中的路网构成变化、区域经济布局变动、区域经济增长速度差异等一些具有时变性和特殊性因素，采用在“四阶段法”运量预测基础上改进而成的“三阶段法”进行高速铁路短期客运量预测。首先采用组合预测模型进行趋势运量预测，然后采用多元LOGIT模型进行方式分担，最后采用弹性系数诱增模型进行诱增运量预测。以京沪高铁为例，采用“三阶段法”预测了2014年和2015年京沪高铁本线及跨线单向客流量。     

Deriving Behavior Model Parameters from Survey Data: Self‐Protective Behavior Adoption During the 2009–2010 Influenza A(H1N1) Pandemic

In this paper, we demonstrate how public opinion surveys can be designed to collect information pertinent to computational behavior modeling, and we present the results of a public opinion and behavior survey conducted during the 2009–2010 H1N1 influenza pandemic. The results are used to parameterize the Health Belief Model of individual health‐protective decision making. Survey subjects were asked questions about their perceptions of the then‐circulating influenza and attitudes towards two personal protective behaviors: vaccination and avoidance of crowds. We empirically address two important issues in applying the Health Belief Model of behavior to computational infectious disease simulation: (1) the factors dynamically influencing the states of the Health Belief Model variables and (2) the appropriateness of the Health Belief Model in describing self‐protective behavior in the context of pandemic influenza.     

论回汉民族关系亲密的四大共因

本文重点对人们司空见惯而又习焉不察的回汉亲密的民族关系的共因作论述和分析。全文依次从回汉血缘关系的共同性,回汉同仇敌忾、反侵略、反压迫的共同性,回汉价值取向和性格的共同性,回汉两族语言使用的共同性四个方面有理有据地证明回汉民族从相互接触到相互学习再到相互认同的过程,正是整个中华民族由分散走向团结再走向和谐凝聚的一个生动缩影。这对了解中华民族多元一体的整合过程,对当今构建和谐社会的和谐民族关系有重要的现实意义。     

全球价值链嵌入与技术创新——基于生产分解模型的分析

基于增加值和最终产品的生产分解模型,本文对我国制造业前向与后向产业关联下的全球价值链嵌入进行测度,实证分析全球价值链嵌入对技术创新的作用机理,并在此基础上重点探讨了吸收能力与技术差距两个重要情境因素的调节作用,同时,运用双重差分、工具变量法以及GMM动态面板模型进行稳健性检验,以控制潜在的内生性问题。研究发现:①我国制造业通过嵌入全球价值链的国际间知识溢出效应促进技术创新能力的提升;②吸收能力能够强化这一正向影响关系;③技术差距在后向全球价值链嵌入对技术创新的影响关系中呈倒U 型调节作用,而在前向全球价值链嵌入对技术创新的影响关系中呈正向调节作用。本文推动了网络嵌入理论和知识溢出理论从组织网络向全球价值链领域的繁衍,丰富了全球价值链嵌入领域的研究成果,同时为我国制造业企业在参与国际分工过程中利用全球价值链嵌入实现技术创新能力提升提供重要的理论参考。     

因子模型门槛效应的LM和Wald检验及其统计性质研究

门槛因子模型可以有效地刻画高维度时间序列的共变特征和区制转换行为,具有良好的可解释性和预测能力。针对因子载荷矩阵存在的门槛效应,本文提出了拉格朗日乘子和沃尔德检验方法,并给出了渐近分布,相关结果表明以上检验统计量具有良好的大样本性质和有限样本表现。在实证部分,以我国股市的行业指数作为研究对象,通过构建门槛因子模型来刻画我国股票市场波动的共变性特征和非对称效应。实证结果表明基于门槛因子模型可以很好地刻画中国股市行业收益率波动的共变特征和区制转换行为。     

大学生村官留任意愿及影响因素分析

利用第一手的大学生村官问卷调查资料,基于Ordered Logistic模型对陕西省282名大学生村官的工作满意度及留任意愿进行了相关分析和回归分析。结果表明：当前大学生村官的工作满意度与留任意愿普遍较低;留任意愿与大学生最初选择从事村官工作的动机紧密相关;大学生村官的性别、年龄、从事村官年限、婚育状况、学历等人口学特征对工作满意度及留任意愿影响显著;社会各界对大学生村官的重视与认可对工作满意度与留任意愿有显著影响;提高大学生村官的收入水平、实行连续三年考核优秀,直接晋升事业编制的政策以及增加晋升为国家公务员等政治机会能够显著提高大学生村官的工作满意度并增强留任意愿,而其它因素对于提高大学生村官的工作满意度与留任意愿的效果并不明显。     

Model Selection and Shrinkage: An Overview

This special issue is concerned with model selection and shrinkage estimators. This Introduction gives an overview of the papers published in this special issue.     

Functional and emotional traits of corporate social media message strategies: Behavioral insights from S&P 500 Facebook data

Drawing from literature regarding public engagement, the Elaboration Likelihood Model (ELM), computer-mediated communication research, and emotion psychology, this study examines the effects of companies’ social media communication strategies on public engagement behaviors as indexed by post likes, shares, and comments. Specifically, it investigates how corporate Facebook posts’ functional traits (functional interactivity and vividness) and emotional traits (emotion presence, valence, and strength) impact public engagement online. Through data mining and computer-assisted sentiment analysis of 33,379 posts from 106 Standard & Poor 500 companies’ Facebook accounts, this study finds a negative effect of functional interactivity but a positive effect of vividness on engagement. It also shows that emotional traits overall yield stronger public engagement outcomes. Two-way interactions between emotional and functional features are also detected. Theoretical and practical implications are discussed.