Adverse Selection in P2P Lending: Does Peer Screening Work Efficiently?—Empirical Evidence from a P2P Platform

The rapid development of online lending in the past decade, while providing convenience and efficiency, also generates large hidden credit risk for the financial system. Will removing financial intermediaries really provide more efficiency to the lending market? This paper used a large dataset with 251,887 loan listings from a pioneer P2P lending platform to investigate the efficiency of the credit-screening mechanism on the P2P lending platform. Our results showed the existence of a TYPE II error in the investors’ decision-making process, which indicated that the investors were predisposed to making inaccurate diagnoses of signals, and gravitated to borrowers with low creditworthiness while inadvertently screening out their counterparts with high creditworthiness. Due to the growing size of the fintech industry, this may pose a systematic risk to the financial system, necessitating regulators’ close attention. Since, investors can better diagnose soft signals, an effective and transparent enlargement of socially related soft information together with a comprehensive and independent credit bureau could mitigate adverse selection in a disintermediation environment.

Reliability and Accuracy of Alternative Default Prediction Models: Evidence from Slovakia

The aim of this paper is to assess the reliability of alternative default prediction models in local conditions, with subsequent comparison with other generally known and globally disseminated default prediction models, such as Altman’s Z-score, Quick Test, Creditworthiness Index, and Taffler’s Model. The comparison was carried out on a sample of 90 companies operating in the Slovak Republic over a period of 3 years (2016, 2017, and 2018) with a narrower focus on three sectors: construction, retail, and tourism, using alternative default prediction models, such as CH-index, G-index, Binkert’s Model, HGN2 Model, M-model, Gulka’s Model, Hurtošová’s Model, Model of Delina and Packová, and Binkert’s Model. To verify the reliability of these models, tests of the significance of statistical hypotheses were used, such as type I and type II error. According to research results, the highest reliability and accuracy was achieved by an alternative local Model of Delina and Packová. The least reliable results within the list of models were reported by the most globally disseminated model, Altman’s Z-score. Significant differences between sectors were identified.

DIFFERENCE OF POWER TEST AND TYPE II ERROR (β) ON MARDIA MVN TEST, HENZE ZIKLER'S MVN TEST, AND ROYSTON'S MVN TEST USING MULTIVARIATE DATA ANALYSIS

The Mardia MVN test, Henze Zikler's MVN test, and Royston's MVN test are the most widely used tests to analyze multivariate normal (MVN) data, but there have not been many studies explaining the advantages and disadvantages of these tests. The research objective was to analyze the difference in test strength and type II (β) error in the Mardia MVN test, Henze Zikler's MVN test, and Royston's MVN test. The research data were analyzed using three MVN tests, namely the Mardia MVN test, Henze Zikler's MVN test, and Royston's MVN test. The results of the analysis in the form of test strength and type II error (β) would be compared at alpha (α) 1%, 5%, 10%, 15%, 20%, and 25%. The comparison results explained that the Mardia test had the greatest test strength and the smallest type II (β) error. The study concluded that the Mardia MVN test was a multivariate normal test better than Henze Zikler's MVN test and Royston's MVN test.

The association between maternal COVID-19 and the rates of cesarean: A systematic review

Background: Maternal COVID-19 infection has been reported to be associated with an increase in the rates of C-sections. However, studies have not reported this association consistently. Therefore, this systematic review and meta-analysis were conducted regarding this issue. Methods: We identified all publications up to January 2021 in databases of PubMed, Scopus, and Web of Science. In addition, we reviewed gray literature and conference proceedings. The pooled odds ratios (OR) and 95% confidence intervals (CI (measured from eligible studies were used as random effect estimates of association among studies. Heterogeneity testing using I2 statistics was performed to assess variance between the studies. Results: The pooled estimates of OR did not report a significant association between COVID-19 and the risk of cesarean section, respectively (OR = 2.02, 95% CI = 0. 4 to 3.64). There was low heterogeneity among the studies reporting the risk of diseases associated with cesarean section among children based on OR (I2=12.7%, P=0.333). The P values for Begg’s and Eggerʼs regression were 0.573 and 0.555, respectively. Conclusion: While we did not find an association between COVID-19 and increased rates of cesarean sections, we cannot exclude a type II error; therefore, additional studies, including large cohort studies by controlling confounding variables, should be performed in the future.

Risk factors for surgical site infection after groin hernia repair: does the mesh or technique matter?

Introduction Surgical site infections (SSIs) following groin hernia repair (GHR) are getting rarer in high income countries despite a wider use of meshes. Among the risk factors for SSIs, those related to the mesh and the surgical technique have rarely been described. Methods A registry-based multicenter study using prospectively collected data, including SSIs and their potential risk factors, was conducted in the French Hernia-Club. Results Between 2012 and 2019, 21,976 consecutive unselected adult patients aged 64.8 ± 15.4 years old (88.9% male) underwent GHR (83.5% unilateral). Fifty four percent were laparoscopic; 97.6% used mesh. The overall incidence of SSI was 0.26%. The incidence of SSI was respectively, 0.24% and 0.19% (p = 0.420) in open vs laparoscopic repairs; 0.19% and 0.25% (p = 0.638) for polyester vs polypropylene mesh; In adjusted multivariate analysis focusing on macroporous meshes (which were the most implanted meshes: 23,148 out of 24,099), there were no differences in terms of SSIs’ rates regarding the technique: open versus laparoscopy (p = 0.762) nor the type of mesh used: polypropylene versus polyester (p = 0.557). Conclusion The rate of SSI following GHR was low in this large registry study. Mesh type and surgical technique did not affect SSIs rates. Caution is advised when interpreting these data due to this very low rate of SSI and the potential for a type II error.

Power Calculator for Detecting Allelic Imbalance Using Hierarchical Bayesian Model

Allelic imbalance (AI) is the differential expression of the two alleles in a diploid. AI can vary between tissues, treatments, and environments. Statistical methods for testing in this area exist, with impacts of explosive type I error in the presence of bias well understood. However, for study design, the more important and understudied problem is the type II error and power. As the biological questions for this type of study explode, and the costs of the technology plummet, what is more important: reads or replicates? How small of an interaction can be detected while keeping the type I error at bay? Here we present a simulation study that demonstrates that the proper model can control type I error below 5% for most scenarios. We find that a minimum of 2400, 480, and 240 allele specific reads divided equally among 12, 5, and 3 replicates is needed to detect a 10%, 20%, and 30%, respectively, deviation from allelic balance in a condition with power >80%. A minimum of 960 and 240 allele specific reads is needed to detect a 20% or 30% difference in AI between conditions with comparable power but these reads need to be divided amongst 8 replicates. Higher numbers of replicates increase power more than adding coverage without affecting type I error. We provide a Python package that enables simulation of AI scenarios and enables individuals to estimate type I error and power in detecting AI and differences in AI between conditions tailored to their own specific study needs.

Distributed Hypothesis Testing over Noisy Broadcast Channels

This paper studies binary hypothesis testing with a single sensor that communicates with two decision centers over a memoryless broadcast channel. The main focus lies on the tradeoff between the two type-II error exponents achievable at the two decision centers. In our proposed scheme, we can partially mitigate this tradeoff when the transmitter has a probability larger than 1/2 to distinguish the alternate hypotheses at the decision centers, i.e., the hypotheses under which the decision centers wish to maximize their error exponents. In the cases where these hypotheses cannot be distinguished at the transmitter (because both decision centers have the same alternative hypothesis or because the transmitter’s observations have the same marginal distribution under both hypotheses), our scheme shows an important tradeoff between the two exponents. The results in this paper thus reinforce the previous conclusions drawn for a setup where communication is over a common noiseless link. Compared to such a noiseless scenario, here, however, we observe that even when the transmitter can distinguish the two hypotheses, a small exponent tradeoff can persist, simply because the noise in the channel prevents the transmitter to perfectly describe its guess of the hypothesis to the two decision centers.

Visual hallucinations in Gentington's disease: ophthalmic validation of ‘North-East Visual Hallucinations Interview

Huntington's disease (HD) is a steadily progressive neurodegenerative disease caused by a mutation in the huntingtin gene, with progressive motor, cognitive and mental impairments.The aim of the study is to determine the validity of the North‑East Visual Hallucinations Interview (NEVHI) in a Russian sample of patients.Materials and methods. 78 subjects were examined: 26 patients at the manifest stage of HD, 21 pre‑manifest carriers of the HD gene and 31 volunteers of the control group.Results. The low incidence of visual hallucinations according to the NEVHI does not allow us to conclude about their specificity for HD. Failure to achieve the level of statistical significance in this case suggests a high risk of type II error associated with the low occurrence of the symptom and the relatively small group size.Conclusion. The study showed that the NEVHI questionnaire can serve as an adequate method for the diagnosis of visual hallucinations in HD.