Bivariate income distributions with lognormal conditionals

In this paper, the most general bivariate distribution with lognormal conditionals is fully characterized, using the methodology proposed by [3]. The properties of the new family are studied in detail, including marginal and conditional distributions, regression functions, dependence measures, moments and inequality measures. The new distribution is very broad, and contains as a particular case the classical bivariate lognormal distribution. Several subfamilies are studied and a generalization of the basic model is discussed. Finally, we present an empirical application. We estimate and compare the basic model proposed in the paper with a classical model, using data from the European Community Household Panel in different periods of time.     

Will They Ever Converge? Earnings of Immigrant and Canadian-born Workers over the Last Two Decades1

Using Census data covering the 1980–2000 period, this article examines what outcomes would be necessary for today's recent immigrant cohorts to achieve earnings parity with Canadian‐born workers. Our results show that today's recent immigrants would have to experience a drastic steepening of their relative age‐earnings profile in the near future for their earnings to converge with their Canadian‐born counterparts. The reason is simple: the greater relative earnings growth experienced by recent immigrant cohorts has only partially offset the drastic deterioration in their relative earnings at entry.     

Emotional displays in children's television programming

The facial expressions of emotion and the circumstances under which the expressions occurred in a sample of the most popular children's television programs were investigated in this study. Fifteen-second randomly selected intervals from episodes of five television programs were analyzed for displays of happiness, sadness, anger, fear, disgust, and surprise. In addition, the contexts in which the emotions occurred were examined. Results indicated that particular emotional expressions occurred at significantly different frequencies and that there was an association between emotional displays and emotion-contexts. The high rate of emotional displays found in television shows has implications for the development of knowledge regarding emotional display rules in viewers.We are grateful to Sharon Galligan for assistance in coding part of the data and to Carolyn Saarni and Amy Halberstadt for helpful comments on an earlier draft of this paper. This research was supported in part by a grant from the National Institute of Disabilities and Rehabilitation Research, #GOO85351. The opinions expressed herein do not necessarily reflect the position or policy of the U.S. Department of Education.     

Control charts for monitoring the median in non-negative asymmetric data

Statistical Methods & Applications - Control charts are commonly used for monitoring the mean of processes. However, there are practical applications in which asymmetric data are the standard....     

Dynamic evolution of alliance portfolios

This research advances in the conceptualization and the operationalization of the factors or microdynamics that determine the choice of partners. Most of the literature in this area of research has centered on the individual alliance without taking into account the set of alliances that a firm makes. This research applies a holistic approach, analyzing the set of alliances as a whole and points to the factors related to alliance portfolios that affect the choice of partners. The authors propose a longitudinal model that allows us to analyze the evolution of interorganizational networks. A Simulation Investigation for Empirical Network Analysis (SIENA) has been applied to a sample of 229 firms, analyzing 10,556 strategic alliances established between them over a 12 year period. The results show that certain factors over which managers exercise a degree of control (deliberate endogenous microdynamics) exert significant influences on the choice of partners by the focal actor and thereby on the dynamic configuration of alliance portfolios.     

Unconstrained parametrizations for variance-covariance matrices

The estimation of variance-covariance matrices through optimization of an objective function, such as a log-likelihood function, is usually a difficult numerical problem. Since the estimates should be positive semi-definite matrices, we must use constrained optimization, or employ a parametrization that enforces this condition. We describe here five different parametrizations for variance-covariance matrices that ensure positive definiteness, thus leaving the estimation problem unconstrained. We compare the parametrizations based on their computational efficiency and statistical interpretability. The results described here are particularly useful in maximum likelihood and restricted maximum likelihood estimation in linear and non-linear mixed-effects models, but are also applicable to other areas of statistics.     

Macroeconomic conditions, institutional factors and demographic structure: What causes welfare caseloads?

Some of the possible determinants of changes observed in welfare caseloads are analysed in this study. The paper attempts to explain variations in welfare caseloads on the basis of four different factors: macroeconomic conditions, interactions with other income maintenance programmes, changes in the programme's parameters and changes in the demographic structure. Different models are tested by applying time series techniques. The data employed are of an administrative nature and cover the whole period that the Madrid programme has been in operation. The results show that the effects of both unemployment and institutional variables are strong and significant. Macroeconomic effects on caseloads could be offset or reinforced by changes in unemployment protection and reforms of the programme's parameters. The significant effects found for the share of lone-parent applicants also show that demographic shifts could shape the trends of welfare rolls regardless of the changes in unemployment or the implementation of institutional reforms. The authors would like to thank Juan Gabriel Rodríguez, Minoru Kunizaki, Miguel Angel Malo and three anonymous referees for their comments and suggestions. Financial and technical support from the Spanish Institute for Fiscal Studies, the Inter-ministerial Commission on Science and Technology (SEC 2001-0746) and the Social Services Department of the Madrid Government is also gratefully acknowledged. Responsible editor: Christian Dustmann.     

Violence Exposure and Drug Use in Central American Youth: Family Cohesion and Parental Monitoring as Protective Factors

Associations between witnessing serious violence and drug use, and the protective influences of family cohesion and parental monitoring, were investigated among 9,840 adolescents (50.5% female, M age=15.29 years, SD=1.76) living in Panama and Costa Rica. After accounting for demographics and parental and sibling substance use, witnessing serious violence was associated with greater drunkenness, tobacco use, number of illicit drugs used, and problems with drugs and alcohol. In every analysis, exposure to violence was associated with more drug use, while family cohesion and parental monitoring attenuated risk. Further, family cohesion and parental monitoring exerted a protective‐stabilizing effect on number of illicit drugs used and on problems with drugs and alcohol. There were few interactions with age. Implications for prevention are discussed.