Costs Of False Positive Errors Research Articles

Specificity and item endorsement rates of personality assessment inventory over-reporting scales across ethnic, gender, and diagnostic groups in a forensic inpatient sample

The accurate evaluation of response style, particularly with respect to overreporting, is imperative in forensic settings wherein an external incentive to feign exists. Given the high cost of false positive errors in this context, as well as the associated cost of false negative errors, evaluators need to ensure that overreporting methods are effective with the unique patient populations with whom they work. Complicating this issue is that forensic samples often differ in predictable ways from the normative samples upon which typical psychological assessment instruments were normed. The purpose of the present investigation was to evaluate the specificity of the overreporting indices on the Personality Assessment Inventory, one of the most commonly used personality inventories, in a forensic sample with no ostensible incentive to feign. Although item endorsement and configural elevations on the Negative Impression Management (NIM) scale and the Malingering Index (MAL) were associated with genuine psychopathology, results indicated that the overall specificity estimates across groups were generally adequate. Further, and consistent with prior research, Rogers Discriminant Function (RDF) performed poorly in this sample.

Probability, Presumptions and Evidentiary Burdens in Antitrust Analysis: Revitalizing the Rule of Reason for Exclusionary Conduct

The conservative critique of antitrust law has been highly influential and has facilitated a transformation of antitrust standards of conduct since the 1970s and led to increasingly more permissive standards of conduct. While these changes have taken many forms, all were influenced by the view that competition law was over-deterrent. Critics relied heavily on the assumption that the durability and costs of false positive errors far exceeded those of false negatives. Many of the assumptions that guided this retrenchment of antitrust rules were mistaken and advances in the law and in economic analysis have rendered them anachronistic, particularly with respect to exclusionary conduct. Continued reliance on what are now exaggerated fears of “false positives,” and failure adequately to consider the harm from “false negatives,” has led courts to impose excessive demands of proof on plaintiffs that belie both established procedural norms and sound economic analysis. The result is not better and more reasonable antitrust standards, but instead an embedded ideological preference for non-intervention that creates a tendency toward false negatives, particularly in modern markets characterized by economies of scale and network effects. In this article, we explain how these erroneous assumptions about markets, institutions, and conduct have distorted the antitrust decision-making process and produced an excessive risk of false negatives in exclusionary conduct cases involving firms attempting to achieve, maintain, or enhance dominance or substantial market power. To redress this imbalance, we integrate modern economic analysis and the teaching of decision theory with the foundational conventions of antitrust law, which has long relied on probability, presumptions, and reasonable inferences to provide more effective means for evaluating competitive effects and resolving antitrust claims.

Effects of misclassification costs on mapping mineral prospectivity

Mineral prospectivity mapping is a classification process because in a given study area, a specific region is classified as either a prospective or non-prospective area. The cost of false negative errors differs from the cost of false positive errors because false positive errors lead to wasting much more financial and material resources, whereas false negative errors result in the loss of mineral deposits. Traditional machine learning algorithms using for mapping mineral prospectivity are aimed to minimize classification errors and ignore the cost-sensitive effects. In this study, the effects of misclassification costs on mapping mineral prospectivity are explored. The cost-sensitive neural network (CSNN) for minimizing misclassification costs is applied to map Fe polymetallic prospectivity in China’s southwestern Fujian metalorganic belt (SFMB). A CSNN with a different cost ratio ranging from 1:10 to 10:1 was used to represent various misclassification costs. The cross-validation results indicated a lower misclassification cost compared to traditional neural networks through a threshold-moving based CSNN. The CSNN’s predictive results were compared to those of a traditional neural network, and the results demonstrate that the CSNN method is useful for mapping mineral prospectivity. The targets can be used to further explore undiscovered deposits in the study area.

An examination of the accuracy of a sequential PCR and sequencing test used to detect the incursion of an invasive species: the case of the red fox in Tasmania

Polymerase chain reaction (PCR) diagnostic tests are increasingly applied to the identification of wildlife. Yet rigorous verification is rare and the estimation of test accuracy (the probability that true positive and true negative samples are correctly identified – test sensitivity and specificity, respectively), particularly in combination with sequencing, is uncommon. This is important because PCR-based tests are prone to contamination in sampling and the laboratory. Here, we use an experimental case–control approach to estimate the sensitivity and specificity of a sequential PCR-based wildlife detection test used to identify incursions of red foxes into Tasmania from predator faeces (scats). Our results show that the sensitivity of the fox test is high (∼94%) for the PCR-based test on its own, but this decreases to ∼84% when combined with the DNA sequencing step. In contrast, the specificity increases from ∼96% in the PCR-only test to ˜99·6% after inclusion of the DNA sequencing step. The intense public scrutiny of the fox eradication programme in Tasmania has undoubtedly influenced the application of a sequential PCR test that maximizes specificity at the expense of sensitivity and so increases the risk that scats containing fox DNA would not be detected. This could lead to the establishment of foxes in Tasmania as a consequence. Synthesis and applications. Importantly, the estimation of the sensitivity and specificity of sequential tests enables decisions about the risk associated with mistaken identification (i.e. false negatives vs. false positives) to be quantified for decision-makers. The cost of false-negative errors should be balanced against the costs of false-positive errors, which could include the expenditure incurred in the application of unnecessary management actions were foxes not in fact present. Understanding the risks and costs associated with both false-negative and false-positive errors is therefore a key component to the decision-making process for the management of the Tasmanian fox incursion.

Iqbal, Twombly, and the Expected Cost of False Positive Errors

Iqbal and Twombly introduced a new standard for pleading federal claims, overruling five-decades old language from Conley v. Gibson. Instead of plaintiffs’ being entitled to discovery unless the complaint affirmatively forecloses the possibility of recovery, Iqbal and Twombly require a more searching evaluation of the complaint under an ambiguous “plausibility” standard. The policy behind this increased burden on plaintiffs is to prevent the false positive error that burdensome discovery creates. How the plausibility standard from Iqbal and Twombly should operate in the real world is poorly understood. There is general acknowledgement that no clear guidance exists about how to ensure only plausible inferences are indulged. Professor Josh Davis recently summed up the sentiments: “Twombly is a very confusing decision. I really do think… it’s going to be a while before we get a rule that produces predictable results.” Professor Adam Steinman is more hyperbolic: “thanks to” Twombly and Iqbal, “federal pleading standards are in a crisis.” This article fills the void in scholarly and judicial thought by providing an understanding of the Twombly/Iqbal pleading standard that is both theoretically sound and pragmatic. Our approach turns on the policy underlying Iqbal and Twombly – a concern that permitting the case to proceed past the motion to dismiss creates a high expected cost of false positive error through non-merits-based settlements. The expected cost of false positive error is a function of two variables: the cost to society if the error occurs and the likelihood that the error will occur. The cost to society if false positive error occurs varies among different areas of substantive law. The likelihood that false positive error will occur varies by the type of facts in controversy. We use antitrust law as an example of the effects of those two variables. Because false positive error in antitrust is costly to society, the analysis focuses on the likelihood of false positive error. The likelihood of false positive error depends on the nature of facts in controversy and the manner in which proof of those facts will be made. If evidence of contested facts is uniquely in the defendant’s hands, it can be retrieved (or its absence shown) only through burdensome, and asymmetrical, discovery of the defendant. If evidence of contested facts is public or quasi-public in nature, it can be gathered without disproportionate burden being placed on the defendant. Whether discovery will be disproportionately burdensome on the defendant will affect the defendant’s incentive to settle. Settlements to avoid burdensome discovery result in false positives. Using antitrust law as an example, we demonstrate that different elements of different substantive claims give rise to different likelihoods of false positives. This approach is trans-substantive, and the same inquiry we conduct should be conducted for any claim brought in federal litigation. This analysis resolves the uncertainty left in the wake of Twombly and Iqbal.