Some results for maximum likelihood estimation of adjusted relative risks

Maximum likelihood (ML) estimation of relative risks via log-binomial regression requires a restricted parameter space. Computation via non linear programming is simple to implement and has high convergence rate. We show that the optimization problem is well posed (convex domain and convex objective) and provide a variance formula along with a methodology for obtaining standard errors and prediction intervals which account for estimates on the boundary of the parameter space. We performed simulations under several scenarios already used in the literature in order to assess the performance of ML and of two other common estimation methods.     

U.S. population in the 21st century: Which scenario is reasonable?

A major Census Bureau study released in January 1989 has evoked renewed warnings in the media and among some population analysts that the U.S. faces population decline in the next century if it does not increase fertility and/or raise immigration. The report's middle scenario rests on an assumed future total fertility rate (TFR) of 1.8, life expectancy of 81.2 years, and net immigration of 500,000 annually. These mid-range assumptions would yield a United States population of 268 million by 2000, peaking at 302 million in 2040 and falling to 292 million by 2080. Questionable assumptions in the report's most likely scenario are discussed. These are:

1. that immigrants bear children at the same rate as their equivalent age and racial group in the United States population.
2. that the high TFR of Hispanics will not raise the overall 1.8 TFR foreseen for whites as the Hispanic proportion of the white population continues to grow.
3. that net yearly immigration will fall to 575,000 in 1990 and 500,000 by 2000. The Census Bureau's "high" assumption of 800,000 net yearly may be more realistic.

The report's low growth scenario projects future population size that is more reassuring than alarming: 264 million in 2000, rising to 288 million in 2030, and falling to 266 million in 2080. Thus, in ninety years the United States would still have 20 million more people than now. While some fear that such slow growth will lower United States influence and bring labor shortages and an aging population, the nation's quality of life would be less at risk with a population of 266 million than with one approaching the one-half billion projected by the Census report's high estimates.     

Predictors for resolution of antisocial behavior among foster care youth receiving community-based services

Youth in the foster care system are more likely to be diagnosed with mental illness than those in the general population. Within this system, youth with antisocial behavior (e.g., aggressive, oppositional) are overrepresented. The challenges youth with antisocial behavior present to foster care systems make understanding the factors that predict remission in this population important for improving placement stability. Using Optimal Data Analysis (ODA), this study examines potential moderating effects of various individual, social, and strength variables on clinically significant decreases antisocial behavior in a sample of foster care youth over time. Results revealed positive improvements in youths' wellbeing to be the optimal predictor of resolution, followed by positive changes in family functioning and positive changes in adjustment to trauma (i.e., symptoms of PTSD). These results indicate that clinically significant decreases over time in antisocial behavior were associated with concurrent improvement in individual and environmental variables. Implications for service providers working with this population are discussed.     

The Shift towards Family Reunification in Romanian Child Welfare Policy: An Analysis of Changing Forms of Governmental Intervention in Romania

This article examines the transition in Romanian child welfare policy from a paternalistic attitude of taking responsibility for ‘abandoned’ children to a capitalist orientation of reinstating responsibility for ‘abandoned’ children to the private sphere. While this shift in child policy is often seen to reflect a withdrawal of the public sphere from the private world of childcare, this study argues, alternatively, that this shift demonstrates a change in the form of governmental intervention. This article is based on ethnographic research that examines the reflections of Romanian child welfare professionals upon their changing roles.     

Sample‐size calculations for multi‐group comparison in population pharmacokinetic experiments

This paper describes an approach for calculating sample size for population pharmacokinetic experiments that involve hypothesis testing based on multi‐group comparison detecting the difference in parameters between groups under mixed‐effects modelling. This approach extends what has been described for generalized linear models and nonlinear population pharmacokinetic models that involve only binary covariates to more complex nonlinear population pharmacokinetic models. The structural nonlinear model is linearized around the random effects to obtain the marginal model and the hypothesis testing involving model parameters is based on Wald's test. This approach provides an efficient and fast method for calculating sample size for hypothesis testing in population pharmacokinetic models. The approach can also handle different design problems such as unequal allocation of subjects to groups and unbalanced sampling times between and within groups. The results obtained following application to a one compartment intravenous bolus dose model that involved three different hypotheses under different scenarios showed good agreement between the power obtained from NONMEM simulations and nominal power. Copyright © 2009 John Wiley & Sons, Ltd.     

地球的电子神经系统——全球感测网

“全球感测网”就像是地球的电子皮肤或电子神经系统一般，由许许多多、各式各样的传感器所连结而成，这些传感器彼此沟通、合作且不眠不休地监测我们的世界。     

Subject‐level matching for imbalance in cluster randomized trials with a small number of clusters

In a cluster randomized controlled trial (RCT), the number of randomized units is typically considerably smaller than in trials where the unit of randomization is the patient. If the number of randomized clusters is small, there is a reasonable chance of baseline imbalance between the experimental and control groups. This imbalance threatens the validity of inferences regarding post‐treatment intervention effects unless an appropriate statistical adjustment is used. Here, we consider application of the propensity score adjustment for cluster RCTs. For the purpose of illustration, we apply the propensity adjustment to a cluster RCT that evaluated an intervention to reduce suicidal ideation and depression. This approach to adjusting imbalance had considerable bearing on the interpretation of results. A simulation study demonstrates that the propensity adjustment reduced well over 90% of the bias seen in unadjusted models for the specifications examined. Copyright © 2013 John Wiley & Sons, Ltd.     

Improved point and confidence interval estimators of mean response in simulation when control variates are used

In discrete event simulation, the method of control variates is often used to reduce the variance of estimation for the mean of the output response. In the present paper, it is shown that when three or more control variates are used, the usual linear regression estimator of the mean response is one of a large class of unbiased estimators, many of which have smaller variance than the usual estimator. In simulation studies using control variates, a confidence interval for the mean response is typically reported as well. Intervals with shorter width have been proposed using control variates in the literature. The present paper however develops confidence intervals which not only have shorter width but also have higher coverage probability than the usual confidence interval     

Configurations of knowledge transfer relations: An empirically based taxonomy and its determinants

We focus in this study on sets of inter-organizational relationships (IORs) by applying a configurational approach that includes both the diversity and the intensity of knowledge transfer IORs. We use a latent class cluster analysis to empirically explore the kinds of IORs configurations. We then use antecedents derived from the IOR literature to explain firm membership in those configurations. Our tests allow us to identify four configurations ranging from isolated innovators to innovating firms embedded in diverse and deep sets of IORs. We show that internal knowledge use by firms, and the types of innovative activities in which they engage, are strong predictors of firm membership in different configurations.