Methods for the Analysis of Explanatory Linear Regression Models with Missing Data Not at Random

Since the work of Little and Rubin (1987) not substantial advances in the analysisof explanatory regression models for incomplete data with missing not at randomhave been achieved, mainly due to the difficulty of verifying the randomness ofthe unknown data. In practice, the analysis of nonrandom missing data is donewith techniques designed for datasets with random or completely random missingdata, as complete case analysis, mean imputation, regression imputation, maximumlikelihood or multiple imputation. However, the data conditions required to minimizethe bias derived from an incorrect analysis have not been fully determined. In thepresent work, several Monte Carlo simulations have been carried out to establishthe best strategy of analysis for random missing data applicable in datasets withnonrandom missing data. The factors involved in simulations are sample size,percentage of missing data, predictive power of the imputation model and existenceof interaction between predictors. The results show that the smallest bias is obtainedwith maximum likelihood and multiple imputation techniques, although with lowpercentages of missing data, absence of interaction and high predictive power ofthe imputation model (frequent data structures in research on child and adolescentpsychopathology) acceptable results are obtained with the simplest regression imputation.     

The Method of Purging Applied to Repeated Cross-Sectional Data

In cross-sectional survey research, it is quite common to estimate the(standardized) effect of independent variable(s) on a dependent variable. However, if repeated cross-sectional data are available, much is to be gained if the consequences of these effects on longitudinal social change are considered.To assess these consequences, we describe a type of simulation in whichlongitudinal shifts in the independent variable's distribution, and longitudinal variation in effect on the dependent variable are `purged' from the data. Although the method of purging is known for many years, we add new practical features by relating the method to logistic and linear regression analysis. Because both logistic and linear regression analysis can be found in all majorstatistical packages, the method of purging is made available to a wider group of social scientists. With the use of repeated cross-sectional data, gathered in the Netherlands between 1970 and 1998, the new practical features of the purging method are shown, using the SPSS package.     

论中国菲利普斯曲线与经济周期阶段

菲利普斯曲线的变动轨迹与经济周期阶段关系密切，随着经济周期阶段的变化，菲利普斯曲线揭示的失业率和通货膨胀率之间的变动关系也会发生相应变化。通过对中国菲利普斯曲线与经济周期阶段关系做理论分析、图形分析、回归分析，可以判断出改革开放以来（准确地说是20世纪80年代初以来），菲利普斯曲线的变动轨迹与经济周期阶段大致吻合，这为判断我国经济周期阶段提供了重要参考，即当前菲利普斯曲线的类型或形状可以为经济周期阶段的判断提供佐证。     

Scale and scope economies of international tourist hotels in Taiwan

Based on the survey data of Taiwan's international tourist hotels in 2000, this paper applies Zellner's seemingly unrelated regression technique to a simultaneous system of a translog multi-product cost function and its corresponding factor share equations to investigate the extents of scale and scope economies of Taiwan's international tourist hotels. The empirical results show that product-specific scale economies exist for accommodation, food and beverage, as well as other services. Ray scale economies are significantly present. Economies of scope exist in providing food and beverage and other services jointly, as well as accommodation and food and beverage services jointly. However, economies of scope may not be present when accommodation and other services are jointly provided.     

日常护理需求的事件数模型

以日常护理需求为实例 ,探讨了应用事件数回归模型所应当注意的事项 ,着重考察 :(1)当事件数变量存在大量 0值时 ,如何在建模时加以处理 ;(2 )如何处理零堆积泊松回归模型中的右屏蔽效应 ;(3)并比较不同统计模型对相同变量的边际效应的影响。最后简单地讨论了零堆积负二项回归模型对我们理解老年人日常护理需求模型的涵义。     

中国制造业比较优势变迁的实证分析

采用独立随机增量过程马尔可夫链的方法对中国149个制造业行业的比较优势动态变迁进行实证分析，旨在反映产业比较优势的演变特征。实证分析结果显示，产业比较优势的各个状态均以较高的概率保持在原有状态，由此反映出中国制造业的比较优势模式和专业化分工模式的演变具有固化性的特征。     

p2: a random effects model with covariates for directed graphs

A random effects model is proposed for the analysis of binary dyadic data that represent a social network or directed graph, using nodal and/or dyadic attributes as covariates. The network structure is reflected by modeling the dependence between the relations to and from the same actor or node. Parameter estimates are proposed that are based on an iterated generalized least-squares procedure. An application is presented to a data set on friendship relations between American lawyers.     

Non‐linear Predictability of Value and Growth Stocks and Economic Activity

Recent empirical evidence suggests that stock market index returns are predictable from a variety of financial and macroeconomic variables. We extend this research by examining value and growth portfolios constructed by book-to-market ratio, and consider whether such predictability is evident here. Further, we assess whether such predictability is better characterised by a non-linear form and whether such non-linear predictability can be exploited to provide superior forecasts to those obtained from a linear model. General non-linearities are examined using non-parametric techniques, which suggest possible threshold behaviour. This leads to estimation of a smooth-transition threshold model, with the results indicating an improved in-sample performance and marginally superior out-of-sample forecast results.     

Locational determinants of major US air passenger markets by metropolitan area

The paper examines the influence of metropolitan characteristics in determining the locations of major air traffic markets in the US. In general, the likelihood of a major air passenger market locating in any given metropolitan area is primarily determined by the metropolitan area's population size and overall propensity for air travel. The study shows that on average, a major air passenger market has over 3 million people while a minor market has a population base of about 760,000. The propensity of a population to fly is mainly a function of a number of social and economic indicators, the most important of which was found to be employment in professional–scientific–technical services and management activities.