Resolving the Mixed Models Controversy

Two standard mixed models with interactions are discussed. When each is viewed in the context of superpopulation models, the mixed models controversy is resolved. The tests suggested by the expected mean squares under the constrained-parameters model are correct for testing the main effects and interactions under both the unconstrained-and constrained-parameters models.     

An exploration of Iranian communication to multiple target audiences

State-controlled media use similar message techniques to target specific publics that counterparts in democratic societies use. We explore talking points (attributes) and themes (frames) through content analysis in state-produced propaganda directed at two different audiences. Domestic and internationally targeted propaganda (N = 1491) from Iran regarding the issue of regional security was reviewed. Results indicated Iran emphasized different attributes and frames based on audiences. Themes about enemy correlated, and relationships between officials and themes were explored.     

Approximate bounded influence estimation for longitudinal data with outliers and measurement errors

Mixed effects models or random effects models are popular for the analysis of longitudinal data. In practice, longitudinal data are often complex since there may be outliers in both the response and the covariates and there may be measurement errors. The likelihood method is a common approach for these problems but it can be computationally very intensive and sometimes may even be computationally infeasible. In this article, we consider approximate robust methods for nonlinear mixed effects models to simultaneously address outliers and measurement errors. The approximate methods are computationally very efficient. We show the consistency and asymptotic normality of the approximate estimates. The methods can also be extended to missing data problems. An example is used to illustrate the methods and a simulation is conducted to evaluate the methods.     

南通蓝印花布的传承、保护和发展及其文化属性

通过研究南通蓝印花布工艺的传承、保护和发展,找出影响其衰落的原因,针对问题,提出南通传统蓝印花布的保护和弘扬策略。通过分析南通蓝印花布的文化属性,得出其基本核心价值之所在,并明确其所具有的两个文化属性：＂非物质文化遗产＂属性和＂传统文化生态＂属性,指出弘扬创新南通蓝印花布的关键是设计创新。     

A bivariate distribution with Lomax and geometric margins

We develop a stochastic model describing the joint distribution of $(X,N)$, where $N$ has a geometric distribution while $X$ is the sum of $N$ dependent, heavy-tail Pareto components. Models of this form arise in many applications, ranging from hydro-climatology to finance and insurance. We present fundamental properties of this vector, including marginal and conditional distributions, moments, representations, and parameter estimation. We also include an example from finance, illustrating modeling potential of this new bivariate distribution.     

The Role of Non‐profit Organizations in the Mixed Economy of Welfare‐to‐Work in the UK and Australia

This article charts the development of welfare‐to‐work policies and compares and contrasts the traditions of delivery in the UK and Australia. We find that in the UK, employment services and social security benefit administration have been dominated by the central state, traditionally affording a key role to civil servants as direct delivery agents. However, in federal Australia, mixed economies of welfare‐to‐work operate in the different states, there is a far greater role for social services and non‐profit organizations are firmly established as key providers of frontline employment services. Since the late 1990s, UK welfare reforms have been gradually following the Australian lead in contracting non‐state actors as delivery agents. As this trend seems set to continue and intensify, we examine the Australian experience in order to reflect on the role of non‐profits in policy reform.     

包含加法模型和混合模型的指数分析体系

统计指数体系分析方法的一个较大不足就是局限于乘法形式的经济方程式,然而经济现象之间的相互联系是多种多样的,目前的统计指数体系分析方法难以应用于形式多样的复杂经济现象的因素分析。本文把基于乘法方式的指数体系称之为指数体系的“乘法模型”,并以此为基础,将传统的只包含乘法关系的经济方程式加以扩展,将其推广到可以应用于加减运算的“加法模型”和包含全部四则运算的“混合模型”,建立了“广义指数分析模型”。该模型不仅保留了传统统计指数体系的基本结构和基本分析方法,还可以较大地提高统计指数分析的深度和广度,由此可以基本解决统计指数体系分析方法的局限性问题。     

On the choice of instruments in mixed frequency specification tests

Abstract

Time averaging has been the traditional approach to handle mixed sampling frequencies. However, it ignores information possibly embedded in high frequency. Mixed data sampling (MIDAS) regression models provide a concise way to utilize the additional information in high-frequency variables. In this paper, we propose a specification test to choose between time averaging and MIDAS models, based on a Durbin-Wu-Hausman test. In particular, a set of instrumental variables is proposed and theoretically validated when the frequency ratio is large. As a result, our method tends to be more powerful than existing methods, as reconfirmed through the simulations.     

Hierarchical Bayes estimation of spatial statistics for rates

The U.S. Bureau of Labour Statistics publishes monthly unemployment rate estimates for its 50 states, the District of Columbia, and all counties, under Current Population Survey. However, the unemployment rate estimates for some states are unreliable due to low sample sizes in these states. Datta et al. (1999) proposed a hierarchical Bayes (HB) method using a time series generalization of a widely used cross-sectional model in small-area estimation. However, the geographical variation is also likely to be important. To have an efficient model, a comprehensive mixed normal model that accounts for the spatial and temporal effects is considered. A HB approach using Markov chain Monte Carlo is used for the analysis of the U.S. state-level unemployment rate estimates for January 2004-December 2007. The sensitivity of such type of analysis to prior assumptions in the Gaussian context is also studied.     

Model-based clustering,classification, and discriminant analysis of data with mixed type

We propose a mixture of latent variables model for the model-based clustering, classification, and discriminant analysis of data comprising variables with mixed type. This approach is a generalization of latent variable analysis, and model fitting is carried out within the expectation-maximization framework. Our approach is outlined and a simulation study conducted to illustrate the effect of sample size and noise on the standard errors and the recovery probabilities for the number of groups. Our modelling methodology is then applied to two real data sets and their clustering and classification performance is discussed. We conclude with discussion and suggestions for future work.