Receiver operating characteristic (ROC) curves are commonly used to evaluate predictions of binary outcomes. When there is a small percentage items interest (as would be the case with fraud detection, for example), ROC can provide an inflated view performance. This cause challenges in determining which set better. In this article, we discuss conditions under precision-recall may preferable curves. As illustrative example, compare two predictors (Beneish’s [1999, Financial Analysts Journal...

Background The global burden of disease in children is large and disproportionally affects low-income middle-income countries (LMICs). Geospatial analysis offers powerful tools to quantify visualise disparities surgical care LMICs. Our study aims analyse the geographical distribution paediatric conditions evaluate access Somaliland. Methods Using Surgeons OverSeas Assessment Surgical Need survey a combined from WHO’s (WHO) Tool—Hospital Walkthrough Global Initiative for Children’s Surgery...

We study the role of reputation in auditor-client matching. Using 1.2 million employment records from US broker-dealers, we find that broker-dealer clients same auditor have similar financial adviser misconduct profiles. Our estimates indicate variation client behavior is nearly half as important size explaining matches. Auditors adjust their portfolios when presented with new information about behavior, and those most significant concerns are least likely to deal high clients. Finally, an...

Abstract This paper presents a procedure for recovering causal coefficients from selected samples that uses random forests, popular machine‐learning algorithm. proposed method makes few assumptions regarding the selection equation and distribution of error terms. Our Monte Carlo results indicate our performs well, even when outcome equations contain same variables, as long is nonlinear. The can also be used there are many variables in equation. We compare with other parametric semiparametric...

A heterogeneous-firm trade model can explain the recent decrease in exchange rate pass-through to aggregate US import prices as a result of decreased costs. This paper finds support for this explanation by testing another implication type heterogeneous firm model: lower goods that are traded short periods time.

Receiver operating characteristic (ROC) curves can be misleading when they are constructed with selected samples. In this article, we describe heckroccurve, which implements a recently developed procedure for plotting ROC The command estimates the area under curve and graphical display of curve. A variety plot options available, including ability to add confidence bands plot.

In this article, we present commands to enable fixing the value of correlation between unobservables in Heckman models. These can solve two practical issues. First, for situations which a valid exclusion restriction is not available, these exploring how results could be affected by sample-selection bias. Second, stepping through values verify whether global maximum likelihood function has been found. We provide several fit and related models with fixed unobservables.

Abstract This paper considers two approaches to examining potential bias in China's consumer price index: (i) inferring true changes cost of living from behaviour; and (ii) creating alternative indices compare with official indices. For (i), our semi‐parametric estimates agree the recent finding a large understatement increases living. (ii), we focus on food prices explore causes bias. We find some evidence new‐good large‐scale migration also that are not reflected index.

Cardiac safety screening is of paramount importance for drug discovery and therapeutics. Therefore, the development novel high-throughput electrophysiological approaches hiPSC-derived cardiomyocyte (hiPSC-CM) preparations much needed efficient testing. Although multielectrode arrays (MEAs) are frequently employed field potential measurements excitable cells, a recent publication by Joshi-Mukherjee colleagues described validated its application recurrent action (AP) recordings from same...

Jessica Ying Chana* & Jonathan A. Cookb*† a United Nations, Department of Economic and Social Affairs, Statistics Division (UNSD)b Public Company Accounting Oversight Board (PCAOB)

ROC curves are commonly used to evaluate predictions of binary outcomes. When there a small percentage items interest (as would be the case with fraud detection, for example), can provide an inflated view performance. This cause challenges in trying determine which set is better. article discusses conditions under precision-recall may preferable curves. As illustrative example, we compare two predictors (Beneish's (1999) M-score and Dechow et al.'s (2011) F-score) using both To aid reader...

This article presents a Stata module to enable fixing the value of correlation between unobservables in Heckman models. These commands can solve two practical issues. First, for situations which valid exclusion restriction is not available, these exploring how results could be affected by sample-selection bias. Second, stepping through values this verify whether global maximum likelihood function has been found. We provide several estimate and related models with fixed unobservables. are...

This paper presents a procedure for recovering causal coefficients from selected samples that uses random forests, popular machine-learning algorithm. proposed method makes few assumptions regarding the selection equation and distribution of error terms. Our Monte Carlo results indicate our performs well, even when outcome equations contain same variables, as long is nonlinear. The can also be used there are many variables in equation. We compare with other parametric semiparametric methods...

Cardiac safety screening is of paramount importance for drug discovery and therapeutics. Therefore, the development novel high-throughput electrophysiological approaches hiPSC-derived cardiomyocyte (hiPSC-CM) preparations much needed efficient testing. Although multielectrode arrays (MEAs) are frequently employed field potential measurements excitable cells, a recent publication by Joshi-Mukherjee colleagues described validated its application recurrent action (AP) recordings from same...

This paper describes strategic behavior in a nonequilibrium model of asset pricing with heterogeneous sophistication. Both risk and return are increasing the na?veté investors market. Optimal investment involves considering effect that na?e have on Further, we derive simple characterization price dynamics results from an arbitrary combination countably infinite set investor types.

Introduction: Greenspace, both residential and public, is associated with improved outcomes for a variety of medical conditions. Its effect on stroke incidence not as well established. In this study, we sought answers to three questions: Does an increase in greenspace, measured by the Normalized Difference Vegetation Index (NDVI), lessen odds stroke? Is socioeconomic status (SES), Area Deprivation (ADI), correlated NDVI? proximity public park risk factor Methods: This matched (1:4)...