Nonparametric covariate adjustment for receiver operating characteristic curves

The accuracy of a diagnostic test is typically characterized using the receiver operating characteristic (ROC) curve. Summarizing indexes such as the area under the ROC curve (AUC) are used to compare different tests as well as to measure the difference between two populations. Often additional information is available on some of the covariates which are known to influence the accuracy of such measures. The authors propose nonparametric methods for covariate adjustment of the AUC. Models with normal errors and possibly non‐normal errors are discussed and analyzed separately. Nonparametric regression is used for estimating mean and variance functions in both scenarios. In the model that relaxes the assumption of normality, the authors propose a covariate‐adjusted Mann–Whitney estimator for AUC estimation which effectively uses available data to construct working samples at any covariate value of interest and is computationally efficient for implementation. This provides a generalization of the Mann–Whitney approach for comparing two populations by taking covariate effects into account. The authors derive asymptotic properties for the AUC estimators in both settings, including asymptotic normality, optimal strong uniform convergence rates and mean squared error (MSE) consistency. The MSE of the AUC estimators was also assessed in smaller samples by simulation. Data from an agricultural study were used to illustrate the methods of analysis. The Canadian Journal of Statistics 38:27–46; 2010 © 2009 Statistical Society of Canada     

Parametric Modeling for Survival with Competing Risks and Masked Failure Causes

We consider a life testing situation in which systems are subject to failure from independent competing risks. Following a failure, immediate (stage-1) procedures are used in an attempt to reach a definitive diagnosis. If these procedures fail to result in a diagnosis, this phenomenon is called masking. Stage-2 procedures, such as failure analysis or autopsy, provide definitive diagnosis for a sample of the masked cases. We show how stage-1 and stage-2 information can be combined to provide statistical inference about (a) survival functions of the individual risks, (b) the proportions of failures associated with individual risks and (c) probability, for a specified masked case, that each of the masked competing risks is responsible for the failure. Our development is based on parametric distributional assumptions and the special case for which the failure times for the competing risks have a Weibull distribution is discussed in detail.     

Evaluation of an upper extremity student-role functioning scale using item response theory

Millions of workers suffer from upper extremity (musculoskeletal) disorders. Many of these workers are predisposed to upper extremity musculoskeletal disorders because of early exposure to ergonomic risks as students. Computer usage for four or more hours remains the greatest risk for upper extremity musculoskeletal disorders for workers and students alike. Developing preventative methods to reduce student exposure, and thus protect future workers from upper extremity musculoskeletal disorders later in life, requires an appropriate measure for upper extremity musculoskeletal disorder prevalence and related limitations for student functioning. Item response theory analysis was used to evaluate and further develop a upper extremity functioning scale for the student role.     

Introduction: moving youth participation forward

Given the emerging interest among researchers, practitioners, and policymakers in youth participation, it is important to examine and assess carefully the promise and challenges of youth engagement.     

Insider ownership and corporate performance: evidence from Germany

In this paper we address the question whether insider ownership affects corporate performance. Evidence from studies dealing with Anglo-Saxon countries is rather inconclusive, especially because results seem to be significantly affected by endogeneity. Economically, this is due to the fact that in these countries insider ownership seems to be mainly driven by management’s compensation contracts. We argue that Germany is different in this regard, as insider ownership is often related to family control, stock-based compensation is less widespread, and the market for corporate control used to be less developed. Starting from this presumption, our data allows an unbiased observation as to whether insider ownership affects firm performance. Using a pooled data set of 648 firm observations for the years 2003 and 1998, we find evidence for a positive and significant relationship between corporate performance—as measured by stock price performance, market-to-book ratio and return on assets—and insider ownership. This relationship seems to be rather robust, even if we account for potential endogeneity by applying a 2SLS regression approach. Furthermore, the results hold for a sub-sample of firms that did not have a stock-based compensation program in place. Moreover, we find outside block ownership as well as more concentrated insider ownership to have a positive impact on corporate performance. Overall, the results indicate that ownership structure might be an important variable explaining the long term value creation in the corporate sector.

Practical Methods for Meeting Remediation Goals at Hazardous Waste Sites

Risk-based cleanup goals or preliminary remediation goals (PRGs) are established at hazardous waste sites when contaminant concentrations in air, soil, surface water, or groundwater exceed specified acceptable risk levels. When derived in accordance with the Environmental Protection Agency's risk assessment guidance, the PRG is intended to represent the average contaminant concentration within an exposure unit area that is left on the site following remediation. The PRG, however, frequently has been used inconsistently at Superfund sites with a number of remediation decisions using the PRG as a not-to-exceed concentration (NTEC). Such misapplications could result in overly conservative and unnecessarily costly remedial actions. The PRG should be applied in remedial actions in the same manner in which it was generated. Statistical methods, such as Bower's Confidence Response Goal, and mathematical methods such as "iterative removal of hot spots," are available to assist in the development of NTECs that ensure the average postremediation contaminant concentration is at or below the PRG. These NTECs can provide the risk manager with a more practical cleanup goal. In addition, an acute PRG can be developed to ensure that contaminant concentrations left on-site following remediation are not so high as to pose an acute or short-term health risk if excessive exposure to small areas of the site should occur. A case study demonstrates cost savings of five to ten times associated with the more scientifically sound use of the PRG as a postremediation site average, and development of a separate NTEC and acute PRG based on the methods referenced in this article.     

Estimation of Error and Bias in Bayesian Monte Carlo Decision Analysis Using the Bootstrap

Bayesian Monte Carlo (BMC) decision analysis adopts a sampling procedure to estimate likelihoods and distributions of outcomes, and then uses that information to calculate the expected performance of alternative strategies, the value of information, and the value of including uncertainty. These decision analysis outputs are therefore subject to sample error. The standard error of each estimate and its bias, if any, can be estimated by the bootstrap procedure. The bootstrap operates by resampling (with replacement) from the original BMC sample, and redoing the decision analysis. Repeating this procedure yields a distribution of decision analysis outputs. The bootstrap approach to estimating the effect of sample error upon BMC analysis is illustrated with a simple value-of-information calculation along with an analysis of a proposed control structure for Lake Erie. The examples show that the outputs of BMC decision analysis can have high levels of sample error and bias.