Variance Estimation in Spatial Regression Using a Non-parametric Semivariogram Based on Residuals

Abstract.  The empirical semivariogram of residuals from a regression model with stationary errors may be used to estimate the covariance structure of the underlying process. For prediction (kriging) the bias of the semivariogram estimate induced by using residuals instead of errors has only a minor effect because the bias is small for small lags. However, for estimating the variance of estimated regression coefficients and of predictions, the bias due to using residuals can be quite substantial. Thus we propose a method for reducing this bias. The adjusted empirical semivariogram is then isotonized and made conditionally negative-definite and used to estimate the variance of estimated regression coefficients in a general estimating equations setup. Simulation results for least squares and robust regression show that the proposed method works well in linear models with stationary correlated errors.     

Infant Exposure Assessment for Breast Milk Dioxins and Furans Derived from Waste Incineration Emissions

Polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) have been detected in human milk samples obtained in several countries. Possible sources include emissions from incineration of municipal waste in resource recovery facilities. A formula is presented for calculating the infant daily dose of dioxin equivalents from breast milk on the basis of the maternal daily intake. Application of the formula suggests that an infant breast-fed for 12 months would receive around 10% of the cumulative exposure dose per body weight that would be received by an adult with 50 years of exposure. Further analysis indicated that the contribution of dioxin equivalents from breast milk to an infant's body concentration at the end of 12 months of breast feeding would amount to 1.7 times the concentration in the mother. However, dioxin and furan emissions from a source calculated to result in worst-case lifetime cancer risks of the order of 1 in 100,000 are only likely to increase breast milk concentrations by around 1%-10% of the levels that have been detected in several countries. This finding suggests that there are major sources of dioxins and furans other than from municipal solid waste incineration that need to be identified.     

Diagnosis Murder: The Death of State Death Taxes

Since 1976, more than 30 states have eliminated their "death" taxes and many others have reduced them. This unexplored case of interstate tax competition presents a unique opportunity to develop a new, more satisfying definition of competitor based on historical elderly migration patterns. Using data from 1967 onward, we outline the recent history of state death tax competitio n and present a spatial econometric analysis. Interstate tax competition is evident and grows stronger when using migration-based definitions of competitors. The article concludes with still more evidence of interstate tax competition—the recent movement by states to effectively revive their death taxes.     

Repeated‐measures models in the analysis of QT interval

Because of the recent regulatory emphasis on issues related to drug‐induced cardiac repolarization that can potentially lead to sudden death, QT interval analysis has received much attention in the clinical trial literature. The analysis of QT data is complicated by the fact that the QT interval is correlated with heart rate and other prognostic factors. Several attempts have been made in the literature to derive an optimal method for correcting the QT interval for heart rate; however the QT correction formulae obtained are not universal because of substantial variability observed across different patient populations. It is demonstrated in this paper that the widely used fixed QT correction formulae do not provide an adequate fit to QT and RR data and bias estimates of treatment effect. It is also shown that QT correction formulae derived from baseline data in clinical trials are likely to lead to Type I error rate inflation. This paper develops a QT interval analysis framework based on repeated‐measures models accomodating the correlation between QT interval and heart rate and the correlation among QT measurements collected over time. The proposed method of QT analysis controls the Type I error rate and is at least as powerful as traditional QT correction methods with respect to detecting drug‐related QT interval prolongation. Copyright © 2003 John Wiley & Sons, Ltd.     

Assessing population differentiation and isolation from single-nucleotide polymorphism data

Summary. We introduce a new, hierarchical, model for single-nucleotide polymorphism allele frequencies in a structured population, which is naturally fitted via Markov chain Monte Carlo methods. There is one parameter for each population, closely analogous to a population-specific version of Wright's F _ST , which can be interpreted as measuring how isolated the relevant population has been. Our model includes the effects of single-nucleotide polymorphism ascertainment and is motivated by population genetics considerations, explicitly in the transient setting after divergence of populations, rather than as the equilibrium of a stochastic model, as is traditionally the case. For the sizes of data set that we consider the method provides good parameter estimates and considerably outperforms estimation methods analogous to those currently used in practice. We apply the method to one new and one existing human data set, each with rather different characteristics—the first consisting of three rather close European populations; the second of four populations taken from across the globe. A novelty of our framework is that the fit of the underlying model can be assessed easily, and these results are encouraging for both data sets analysed. Our analysis suggests that Iceland is more differentiated than the other two European populations (France and Utah), a finding which is consistent with the historical record, but not obvious from comparisons of simple summary statistics.     

Disability, gender, and unemployment relationships in the United States from the behavioral risk factor surveillance system

Women with disabilities face simultaneous oppression in employment due to discrimination with regard to disability and gender. This article investigates the potential disparity in participation in employment for women, particularly women with disabilities. We analysed weighted data from disability surveillance programs and the Behavioural Risk Factor Surveillance System (BRFSS) on over 47,000 respondents. The disability BRFSS was a telephone survey in 11 states and Washington DC. Logistic regression analyses produced adjusted models of the association between gender and employment. Compared with people without disabilities, there were disparities found for people with disabilities, and women with and without disabilities, with the larger discrepancy for women without disabilities. Additional detail about level of employment is needed to make conclusive statements; however, it is clear that disparities in employment continue to exist for women, regardless of their disability status.     

Childhood abuse and psychiatric impairment in a sample of welfare to work women

Since 1996, individuals who participate in welfare programs are mandated to find employment. Welfare recipients may have difficulty transitioning to the work force due to impairments in psychosocial functioning. Examples include mental health and substance use disorders, medical problems, academic and learning difficulties, and lack of resources such as childcare and transportation. In addition, many welfare recipients have histories of abuse and violence, which further complicate their job finding efforts given the documented relationship between past maltreatment and psychiatric impairment. An area unexplored is the relationship between a history of childhood abuse and psychiatric impairment in Welfare to Work (WtW) recipients. We evaluated the relationships among demographics, personality pathology, substance abuse, cognitive functioning, and histories of childhood abuse in a group of 158 WtW recipients in order to assess predictors of psychiatric impairment. Results suggest that participants with higher levels of psychiatric impairment also had higher levels of interpersonal problems related to personality pathology, higher drug and alcohol use impairment, and endorsed more severe histories of childhood abuse. Personality pathology, alcohol use severity, and self-reported childhood abuse emerged as the best predictor of psychiatric impairment in this WtW sample. These results have implications for identifying individuals at risk for unsuccessful transition into the work force and for the development of effective rehabilitation strategies that considers the unique needs of women in WtW programs.     

Social Networks and Unobserved Heterogeneity in Risk for AIDS

As a sexually transmitted disease, AIDS spreads along social networks; consequently,it is reasonable to propose to utilize these networks in teaching people to avoid practices that put them at increased risk of contracting AIDS. Most obviously, homosexual men are both at relatively high risk of contracting AIDS, and in many urban areas have well crystallized community structures and high social connectivity. We present evidence suggesting that using such social networks can have the unanticipated consequence of reaching a set of men who are at relatively low risk. Evidently, there is great unobserved heterogeneity among the population in terms of risk, and while this heterogeneity is not captured by conventional measures of risk behavior, it is closely linked to network processes.