Discrete time modelling of disease incidence time series by using Markov chain Monte Carlo methods

Summary.  A stochastic discrete time version of the susceptible–infected–recovered model for infectious diseases is developed. Disease is transmitted within and between communities when infected and susceptible individuals interact. Markov chain Monte Carlo methods are used to make inference about these unobserved populations and the unknown parameters of interest. The algorithm is designed specifically for modelling time series of reported measles cases although it can be adapted for other infectious diseases with permanent immunity. The application to observed measles incidence series motivates extensions to incorporate age structure as well as spatial epidemic coupling between communities.     

Biosecurity Policy and the Use of Geospatial Predictive Tools to Address Invasive Plants: Updating the Risk Analysis Toolbox

International and national biosecurity policies consider risk assessment a critical component of overall plant health risk analysis. The Agreement on the Application of Sanitary and Phytosanitary Measures, the International Plant Protection Convention, and the Convention on Biological Diversity all provide guidelines and recommendations on how to use risk assessment. This article discusses how these instruments address risk assessment, and makes recommendations on how the risk assessment process needs to incorporate current geospatial predictive science and geographic information systems into the plant health biosecurity risk analysis toolbox.     

Cognitive skills and leadership performance: The nine critical skills

Over the years, a growing body of evidence indicates that certain cognitive skills are a critical determinant of leader performance. In the present effort we show that application of these skills is typically founded in case-based knowledge. Subsequently, we identify nine key skills people in leadership positions employ when working with case-based knowledge to address leadership problems: 1) problem definition, 2) cause/goal analysis, 3) constraint analysis, 4) planning, 5) forecasting, 6) creative thinking, 7) idea evaluation, 8) wisdom, and 9) sensemaking/visioning. Individual and situational contingencies shaping effective application of these thinking skills is discussed. In addition, the implications of findings bearing on leader thinking skills for leader assessment and leader development are considered.     

Imputation procedures for categorical data: their effects on the goodness-of-fit chi-square statistic

An imputation procedure is a procedure by which each missing value in a data set is replaced (imputed) by an observed value using a predetermined resampling procedure. The distribution of a statistic computed from a data set consisting of observed and imputed values, called a completed data set, is affecwd by the imputation procedure used. In a Monte Carlo experiment, three imputation procedures are compared with respect to the empirical behavior of the goodness-of- fit chi-square statistic computed from a completed data set. The results show that each imputation procedure affects the distribution of the goodness-of-fit chi-square statistic in 3. different manner. However, when the empirical behavior of the goodness-of-fit chi-square statistic is compared u, its appropriate asymptotic distribution, there are no substantial differences between these imputation procedures.     

Queer-friendly nation? The experience of Chinese gay immigrants in Canada

In the past decade, the People’s Republic of China has become the leading source country of immigrants to Canada. However, little is known about the experience of Chinese gay immigrants in Canada. Using narratives collected through a qualitative study, we show the way in which Chinese gay men construct their experience of immigration in Canada. Unlike the national discourse that claims Canada is a friendly nation for members of the LGBTTQ community, the experiences of Chinese gay immigration present a different story. As racialised gay men, they continue to be subjected to social violence in Canada – albeit differently from what they experienced in China. The analysis demonstrates the complexity of social marginalisation experienced by these men, the understanding of which is essential for social workers to engage in practice with them that is socially transformative.     

Finding the Delicate Balance: Serials Assessment at the University of Nevada,Las Vegas

Abstract

In the spring of 2003 the University Libraries at University of Nevada, Las Vegas (UNLV), conducted a Serials Assessment Project which combined technology with effective communication between library users and subject librarians to successfully identify journals that could be cancelled, should cancellations be necessary due to budgetary constraints. Although the library had conducted similar assessments in 1998 and 2000, a fresh look at past assessments resulted in an innovative approach. The project and experience proved to be tremendously positive for both the library and its users. This article describes the planning and process by which the UNLV libraries designed and conducted the 2003 review of UNLV's serials collection, comparing it to methods used in the two previous assessments.     

Socio-economic effects of relative income and relative cohort size

A nonlinear three-stage least-squares technique is used to estimate a four-equation model in which age-specific fertility, marriage, divorce, and female labor force participation rates are the dependent variables. In addition to a range of explanatory variables, two proxies for the Easterlin hypothesis are tested within the model, one measuring relative income and the other measuring relative cohort size. Findings indicate that the Easterlin proxies, while not statistically significant in all cases, help to explain time-series movements in various socio-economic variables in the postwar United States.     

ON THE EFFICIENCY OF THE QUASI-LIKELIHOOD ESTIMATORS FOR EXPONENTIAL FAMILIES WITH EXTRA VARIATION

Bounds are obtained for the asymptotic efficiency of the quasi-likelihood estimator relative to the likelihood estimators for exponential families with random variables in the exponent. We conclude that the quasi-likelihood is very efficient in the Poisson case for small means or small extra variation.