Rare Events Logistic Regression Research Articles

Determinants of Exchange Rate Jumps in Türkiye

This study examines the exchange rate jumps in Türkiye between 2013 and 2021 and the factors influencing these jumps. The Turkish Lira experienced a consistent depreciation against other currencies during this period. A closer examination of the depreciation timeline revealed that the Turkish Lira's depreciation was occasionally abrupt and exceedingly pronounced. The primary objective of this research is to identify these episodes of spikes, which can be characterized as jumps in exchange rates amid regular increases. To achieve this, the Pruned Exact Linear Time (PELT) algorithm was employed to detect sudden shifts in the exchange rate. Taking these points as dependent variables, a rare event logistic regression model was utilized to determine the probability of an exchange rate jump. The findings indicate that increased dollarization raises the likelihood of an exchange rate jump, while higher deposit rates and central bank reserves reduce the probability of a jump.

Rare events model of the MBSAQIP database: risk of early bowel obstruction following metabolic surgery

BackgroundEarly small bowel obstruction (eSBO) (within 30-days) is a rare but important complication that is associated with high rates of morbidity, including readmission, reintervention, and reoperation. ObjectivesTo identify patient-specific and operation-specific characteristics that predispose patients to eSBO and to identify at-risk individuals preoperatively. Setting2015-2021 Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP). MethodsUtilizing the 2015-2021 MBSAQIP PUF, 1,016,484 records were analyzed. Pediatric, revisional, open-conversion, and cases with incomplete data in sex, body mass index, operative-time, 30-day-follow-up variables were excluded. Case details were compared using Fisher’s exact & Wilcoxon -Mann -Whitney tests to identify at-risk patients. The likelihood of eSBO was modeled with rare event logistic regression. ResultsIncidence of eSBO was .40%. Of the 4103 occurrences of eSBO, RYGB (Roux-en-Y gastric bypass), SG (sleeve gastrectomy), and DS (duodenal switch) accounted for 79.4%, 19.3%, and 1.3%, respectively. Many patient-specific characteristics were significantly associated with eSBO. History of prior foregut surgery, a non-metabolic surgery trained operator, and longer operative times were all associated with increased eSBO (P < .0001). While simultaneously controlling for these factors, eSBO remained higher in DS (OR 9.55, P < .0001) and RYGB (OR 5.18, P < .0001) compared to SG. Increased length of operation (OR 1.03, P < .0001) and non -MS-trained operators (OR 1.33, P < .0001) remained highly significant. Male-sex (OR .70, P < .0001) and diabetes (OR .78, P < .0001) were both protective. ConclusionsIn the largest analysis to date, eSBO remains a rare event. RYGB accounts for the largest proportion of eSBO, however, DS has a higher risk adjusted rate of eSBO.

Immigrant Threat or Institutional Context? Examining Police Agency and County Context and the Implementation of the 287(g) Program

ABSTRACT During the early 2000s, the U.S. government began to partner with local law enforcement agencies for assistance with the enforcement of immigration laws. Participation in the voluntary 287(g) program by law enforcement agencies was an important early iteration of this new strategy. Although relatively few agencies adopted 287(g), its implementation was a critical turning point in the nation’s approach to immigration enforcement and has been linked to a wide range of negative consequences for Latinos, including higher levels of violent victimization, and decreases in police trust. This study advances knowledge about the factors associated with 287(g) program adoption during the late 2000s. We examined the role of agency ethnoracial diversity and fiscal capacity and several theoretically relevant indicators of the broader county context. Based on a sample of 880 law enforcement agencies located in 233 counties, rare event logistic regression models indicate that the strongest predictors of 287(g) adoption were the share of county population estimated to be undocumented immigrants, the degree of support for the Republican Presidential candidate, and the relative unemployment of Hispanic and non-Hispanic White residents. Overall, the results suggest that 287(g) was primarily a political response to concerns about the potential threats associated with undocumented immigration.

Examining divorce risk through gender roles in Pakistan

This research investigates the risk of divorce in Pakistan, considering various aspects of gender roles. We used rare events logistic regression to determine the risk of divorce based on socioeconomic variables. We find that females are more likely to be divorced than their male counterparts in all estimation scenarios. While our analysis partially supports the gender role specialization theory, both males and females who do not participate in the labor market are less likely to divorce according to empirical analysis. We also found that people living in urban areas are more likely to divorce due to a lack of social integration. The risk of divorce in various occupations remained the same across years of education and rural/urban living. Our results show that years of education decrease the likelihood of divorce. The current study fills the literature gap by analyzing divorce risk in Pakistan using several theories and hypotheses, most of which are based on gender roles. The study provides a significant number of observations and data on both married and divorced individuals, enabling a thorough investigation of divorce risk in Pakistan.

Fatal crashes and rare events logistic regression: an exploratory empirical study.

Fatal road accidents are statistically rare, posing challenges for accurate estimation through the classic logit model (LM). This study seeks to validate the efficacy of a rare events logistic model (RELM) in enhancing the precision of fatal crash estimations. Both LM and RELM were employed to examine the relationship between pertinent risk factors and the incidence of fatal crashes. Crash-injury datasets sourced from Hillsborough County, Florida served as the empirical basis for evaluating the performance metrics of both LM and RELM. The analysis revealed that RELM yielded more accurate predictions of fatal crashes compared to LM. Receiver operating characteristic (ROC) curves were constructed, and the area under the curve (AUC) for each model was computed to offer a comparative performance assessment. The empirical evidence notably favored RELM over LM as substantiated by superior AUC values. The study offers empirical validation that RELM is demonstrably more proficient in predicting fatal crashes than the LM, thereby recommending its application for nuanced traffic safety analytics.

Loneliness and firearm acquisition among low-income U.S. veterans: Findings from the National Veteran Homeless and Other Poverty Experiences (NV-HOPE) study

RationaleThere are theoretical reasons to expect loneliness related to firearm purchasing. For instance, loneliness might amplify social isolation and feelings of insecurity, anxiety, perceived danger, and need for self-reliance. Purchasing a firearm might be seen as a way to bolster one's security and gain a sense of control when faced with potential threats, yet there is a lack of research assessing this possibility. ObjectiveThis study aimed to examine the relationship between loneliness and purchasing a firearm among a sample of low-income U.S. military veterans. MethodsData are from a national survey of 1,004 low-income U.S. veterans collected in December 2022 and January 2023. Firth logistic regression—a rare event logistic regression model to address small-sample bias stemming from rare outcomes through a penalized likelihood approach—was used to estimate the adjusted association between loneliness and purchasing a firearm in the past year. ResultsApproximately 5.4% reported the purchase of a new firearm in the past year, and respondents reported an average loneliness score of 1.82 (standard deviation = 1.96; range = 0–6). Results of Firth Logit regression analyses found that net of control variables for demographic, socioeconomic, and household characteristics, loneliness was positively associated with purchasing a firearm in the past year (OR = 1.167, 95% CI = 1.006, 1.352). ConclusionThe findings indicate that loneliness among U.S. military veterans is associated with firearm purchasing. Considering the risks posed by both loneliness and firearm ownership for veteran suicide, the findings suggest the need for greater focus and synergistic strategies to address loneliness and safe firearm practices among U.S. veterans.

Large long-haul aircraft entry, survival, and exit in China's air transport market: The Airbus 380 and Boeing 747-8I

This study explores the dynamics of entry, survival, and exit of large long-haul aircraft in China's burgeoning air transport market, focusing on the Airbus 380 and Boeing 747–8I as case studies. Utilizing rare events logistic regression and the time-dependent Cox regression model, the determinants of route entrance and survival for these aircraft from 2011 to 2021 were analyzed. The data is based on the flight plans of pertinent air routes and aircraft. Key findings include: (1) Both aircraft types favored entry into home-airfield-related, long-haul routes with a substantial existing flight volume. (2) Both aircraft exhibited prolonged survival on international long-haul routes, especially when airlines used them to bolster market share with consistent frequency. (3) Without the home-airfield advantage enjoyed by Air China, China Southern Airlines trialed the Airbus 380 on multiple occasions. Notably, if the Airbus 380 maintains operation on a route for at least 16 weeks, it demonstrates a lower likelihood of exit compared to the Boeing 747–8I.

Environmental, climatic, and situational factors influencing the probability of fatality or injury occurrence in flash flooding: a rare event logistic regression predictive model.

Flash flooding is considered one of the most lethal natural hazards in the USA as measured by the ratio of fatalities to people affected. However, the occurrence of injuries and fatali- ties during flash flooding was found to be rare (about 2% occurrence rate) based on our analysis of 6,065 flash flood events that occurred in Texas over a 15-year period (2005 to 2019). This article identifies climatic, environmental, and situational factors that affect the occurrence of fatalities and injuries in flash flood events and provides a predictive model to estimate the likelihood of these occurrences. Due to the highly imbalanced dataset, three forms of logit models were investigated to achieve unbiased estimations of the model coef- ficients. The rare event logistic regression (Relogit) model was found to be the most suit- able model. The model considers ten independent situational, climatic, and environmental variables that could affect human safety in flash flood events. Vehicle-related activities dur- ing flash flooding exhibited the greatest effect on the probability of human harm occur- rence, followed by the event's time (daytime vs. nighttime), precipitation amount, location with respect to the flash flood alley, median age of structures in the community, low water crossing density, and event duration. The application of the developed model as a simula- tion tool for informing flash flood mitigation planning was demonstrated in two study cases in Texas.

HOW DOES THE WTO DISPUTE SETTLEMENT MECHANISM WORK FOR WTO MEMBERS? A COMPARATIVE ANALYSIS OF DEVELOPED AND DEVELOPING COUNTRIES

The WTO dispute settlement mechanism (DSM) is now facing a crisis of paralysis. Given such a challenging environment, this study examines factors affecting the participation of the WTO DSM by using bilateral data of WTO members from 1995 to 2017. Moreover, we provide a comparative analysis of the factors affecting the filing of cases at the WTO DSM between developed and developing countries across this period. By conducting the rare-event logistic regression method, we find that the export intensity, retaliation capability, economic power and economic threat are the main factors determining the initiation of trade disputes by WTO members. Moreover, the results from seemingly unrelated regressions suggest that economic power, proxied by the complainant country’s gross national income, is not a vital issue of importance that developed countries need to consider when deciding to initiate trade disputes; however, in contrast, developing countries still see it as an important factor affecting the use of DSMs, especially when the target countries are developed countries. Nonetheless, it is worth noting that in the later stage of the WTO, the impact of economic power no longer has any differential influence. On balance, these results reflect the efforts of the WTO to build a fair and rule-based DSM while also highlighting the urgent practical significance of maintaining a solid and effective mechanism for handling international trade disputes.

Students and protests: A quantitative cross-national analysis

Previous studies have found a positive relationship between the youth and the educated with protest number, but the form that these protests take needs further research. We argue that students are a unique group, acting neither as an educated nor a young population, and three possible mechanisms push students toward non-violent rather than violent forms of protest. By promoting values of tolerance, higher levels of human capital, and social mobility, education serves as a factor that pacifies destructive tendencies in protest movements. At the same time, universities are a platform for cooperation, and the large amounts of free time and energy make the costs of participating in protests for students minimal compared with other groups. Using a negative binomial regression and a rare events logistic regression, we find that the proportion of students is a strong and consistently significant predictor of the number of nonviolent demonstrations. However, the share of students in the total population does not turn out to be significantly associated with violent protests/armed uprisings.

Vicious Cycle: Violations of Foreign Nationals’ Rights among CAT Countries

ABSTRACTThis article explains abuse of foreigners’ rights by identifying an unexplored by-product of human rights shaming—reciprocation. By identifying this phenomenon, the article explains countries’ noncompliance with treaty obligations, focusing in particular on noncompliance with the Convention against Torture's (CAT) provisions on foreign nationals’ rights. States who learn about violations of their citizens’ rights in other ratifying countries violate foreigners’ rights at home. The article hypothesizes that the likelihood of violations of foreign nationals’ rights in a country is positively associated with mentions of violations of that country's citizens’ rights in other treaty countries. To test these propositions, the article presents new data on countries’ violations of foreign nationals’ human rights as documented in the CAT Treaty Body's reports. The results of rare events logistic regression analysis support the hypothesis. Conceptualizing this dynamic among CAT countries as negative diffuse reciprocity, the article draws theoretical conclusions and policy implications.

Association of progressive optic disc tilt with development of retinal nerve fibre layer defect in children with large cup-to-disc ratio

Background/aimsWhereas myopic optic disc deformation has been posited as a risk factor for glaucomatous damage, longitudinal studies evaluating their association have been sparse. We investigated whether the optic nerve head...

Muscle strength is associated with COVID-19 hospitalization in adults 50 years of age or older.

BackgroundWeak muscle strength has been associated with a wide range of adverse health outcomes. Yet, whether individuals with weaker muscle strength are more at risk for hospitalization due to severe COVID‐19 is still unclear. The objective of this study was to investigate the independent association between muscle strength and COVID‐19 hospitalization.MethodsData from adults 50 years of age or older were analysed using logistic models adjusted for several chronic conditions, body‐mass index, age, and sex. Hand‐grip strength was repeatedly measured between 2004 and 2017 using a handheld dynamometer. COVID‐19 hospitalization during the lockdown was self‐reported in summer 2020 and was used as an indicator of COVID‐19 severity.ResultsThe study was based on the Survey of Health, Ageing and Retirement in Europe (SHARE) and included 3600 older adults (68.8 ± 8.8 years, 2044 female), among whom 316 were tested positive for the severe acute respiratory syndrome coronavirus 2 (8.8%), and 83 (2.3%) were hospitalized due to COVID‐19. Results showed that higher grip strength was associated with a lower risk of COVID‐19 hospitalization [adjusted odds ratio (OR) per increase of 1 standard deviation in grip strength = 0.64, 95% confidence interval (95% CI) = 0.45–0.87, P = 0.015]. Results also showed that age (OR for a 10 ‐year period = 1.70, 95% CI = 1.32–2.20, P < 0.001) and obesity (OR = 2.01, 95% CI = 1.00–3.69, P = 0.025) were associated with higher risk of COVID‐19 hospitalization. Sensitivity analyses using different measurements of grip strength as well as robustness analyses based on rare‐events logistic regression and a different sample of participants (i.e. COVID‐19 patients) were consistent with the main results.ConclusionsMuscle strength is an independent risk factor for COVID‐19 severity in adults 50 years of age or older.

Performance of prior and weighting bias correction methods for rare event logistic regression under the influence of sampling bias

The problem of classifying events to binary classes has been popularly addressed by Logistic Regression Analysis. However, there may be situations where the most interested class of event is rare such as an infectious disease, earthquake, financial crisis etc. The model of such events tends to focus on the majority class, resulting in the underestimation of probabilities for the rare class. Additionally, the model may incorporate sampling bias if the rare class of the sample is not representative of its population. It is therefore important to investigate whether such rareness is genuine or caused by an improperly drawn sample. We conducted a simulation study by creating three populations with different rarity levels and drawing samples from each of those which are either compatible or incompatible with the actual rare classes of the population. Then, the effect of sampling bias is discussed under the two correction methods of bias due to rareness as suggested by King and Zeng.

281. Detecting bacterial sepsis among allogeneic HCT recipients with population-specific bedside tools

BackgroundDiagnosing sepsis among allogeneic hematopoietic cell transplant (aHCT) recipients remains challenging. Existing criteria, for use in hospitalized patients, have limited predictive accuracy among aHCT recipients and their use may lead to missed events or antibiotic overuse. We developed bedside bacterial sepsis prediction tools (criteria and decision tree [DT]) for aHCT recipients and compared them against Systemic Inflammatory Response Syndrome (SIRS), quick Sequential Organ Failure Assessment (qSOFA) and National Early Warning Score (NEWS) criteria.MethodsAdult aHCT recipients transplanted between September 2010–2019 with ≥ 1 potential infection (PI) within 100 days post-transplantation were randomly assigned to model/validation (7/3) cohorts. Tools included demographic and clinical factors and were built against a bacterial sepsis endpoint (gram-negative, Staphylococcus aureus, or Streptococcus species bacteremia). The tools were developed using best subset selection with rare event logistic regression (criteria) and classification tree (DT) algorithms. Criteria scores were estimated using a beta/10 integer weighting approach and tool predictive performances were compared against existing criteria.ResultsBetween September 2010–2019, 1571 recipients with ≥ 1 PI contributed 7755 PIs and 238 sepsis events. The DT model included 7 terminal nodes based on 3 predictors: temperature, respiratory rate (RR), and sex. The criteria model contained 10 categories with 4 predictors: RR, temperature, pulse, and diastolic blood pressure (Figure 1). Our criteria and DT had AUCs of 71.1% (95% Confidence Interval (CI): 64.3, 77.9%) and 70.0% (CI: 63.7, 76.2%). SIRS had the highest AUC of existing criteria – 64.7% (CI: 57.1, 71.9%). Our criteria had the highest net benefit (for probabilities < 10%) and, at a 7+ cut-point, had a sensitivity of 73.8% (CI: 61.5–84.0%) and specificity of 55.0% (CI: 52.9, 57.1%) (Figure 2). ConclusionWe developed aHCT recipient-specific bedside bacterial sepsis prediction tools with higher AUCs than existing criteria. Tools targeted to high-risk populations may lead to fewer missed sepsis events and, in turn, reduce sepsis related mortality among this high-risk population.Disclosures Steven A. Pergam, MD, MPH, Chimerix, Inc (Scientific Research Study Investigator)Global Life Technologies, Inc. (Research Grant or Support)Merck & Co. (Scientific Research Study Investigator)Sanofi-Aventis (Other Financial or Material Support, Participate in clinical trial sponsored by NIAID (U01-AI132004); vaccines for this trial are provided by Sanofi-Aventis)

Who may benefit from lower dosages of intravenous tissue plasminogen activator? Results from a cluster data analysis

BackgroundThe risk of symptomatic intracranial haemorrhage (sICH) after thrombolysis is low but severe. Lower dose of alteplase may reduce the risk of sICH. We aim to identify subsets of patients...

The Sustainability of the WTO Dispute Settlement System: Does It Work for Developing Countries?

Our study provides an econometric analysis of whether developing countries are disadvantaged by theWorld Trade Organization’s (WTO’s) dispute settlement procedure. Despite the many successes of the Dispute Settlement Board (DSB), there is a need to better understand the overall effectiveness of the dispute mechanism. Using bilateral data of WTO members from 1995 to 2017, this article describes the influencing factors for developing countries’ participation in the dispute settlement process. By conducting a rare events logistic regression analysis, the results from this study show that the key factors that determine the ability for a developing country to initiate a dispute against another country include the export intensity, retaliation capability, economic power, and economic threat from potential defendant countries. This means that not all developing countries can participate in trade dispute settlement mechanisms, so the WTO dispute settlement system has relative legitimacy. In addition, we further compare and analyse the influence of these four dimensions over different time periods and in different target countries through seemingly unrelated estimations. WTO, dispute settlement system, developing country

Murder as Importation: An Empirical Study of the Continuity of Homicide Offending From Community to Confinement

Prison murder is the most severe form of institutional violence but its exceedingly low prevalence has limited prior research. Recent studies of prison murders make clear that serious, violent, and chronic career criminals are most likely to perpetrate inmate murders with equivocal evidence of the role of prior homicide offending on prison murder. Using retrospective administrative data from 1,005 prisoners selected from the southwestern United States, the current study examined whether homicide offending in the community is itself an importation factor that is useful for understanding prison murder and thus can be used to understand continuity in homicidal offending from the community to confinement context. Rare events logistic regression models found that individuals sentenced for first-degree murder are more likely to perpetrate prison murder. A separate rare events logistic regression model with any type of homicide commitment offense as a predictor provided similar findings suggesting these effects are robust to model specification. Given its gravity and fundamental threat to prison safety and security, we encourage data collection and additional research on prison murder and the inmates that perpetrate it.

Incidence of Glaucoma or Ocular Hypertension After Repeated Anti-Vascular Endothelial Growth Factor Injections for Macular Degeneration.

PurposeTo estimate the risk of glaucoma or sustained ocular hypertension (OHT) related to anti-vascular endothelial growth factor (VEGF) injections for age-related macular degeneration (AMD).DesignRetrospective chart review.SubjectsPatients who received unilateral anti-VEGF injections for AMD at the Wheaton Eye Clinic (IL).MethodsChart analysis was performed on 1095 patients, without prior glaucoma or OHT, who received unilateral anti-VEGF injections for AMD from 2005 to 2012, with data collected through 2013. Data collection included demographics, lens status, date and medication type of each injection, and the date of diagnosis of glaucoma or OHT by a treating glaucoma specialist, which was the main outcome measure. Rare events logistic regression was performed to determine the risk of disease development based on sex, lens status, and injection frequency.ResultsUnilateral glaucoma or sustained OHT developed in 42 patients over the course of follow-up, with 40 events in the injected eye only, 2 in the contralateral eye only. Statistical modeling predicted elevated risk for onset of glaucomatous disease with a higher maximum frequency of injections (p < 0.0001, odds ratio [OR] 2.18 for each additional injection over the most injection-intense 6 months for a given subject) and with phakic lens status (p = 0.0009, OR 0.33 for pseudophakia).ConclusionOur results show a significant risk for glaucoma or OHT development in patients undergoing repeated treatments with intravitreal anti-VEGF injections for AMD, establishing the first reliable connection between disease development and a period of high-frequency injections. In addition, we show a significantly increased risk of disease development in phakic patients, which we believe points to a mechanical explanation for this type of secondary glaucoma.

A Statistical Risk Assessment Framework for Distribution Network Resilience

Due to the rapid development of distributed renewable generation, an effective risk assessment and early warning mechanism for active distribution networks is of great significance to maintain the system reliability and enhance energy grid resilience. In this paper, a novel risk assessment model is proposed to assess the probability of potential disturbances to the grid and provide accurate advice for trading prosumers' renewable energy. The model can compute node failure probability (FP) for transmission networks as well as the area FP for distribution networks, while combining the two perspectives by topology analysis. A weather threshold value is first derived to define the extreme weather condition. Then the FP of transmission lines is calculated by joint probability models under four instances of extreme climate. For distribution networks, the weather influence is obtained by applying the Rare Events Logistic Regression model initially. Then the equipment fault related to the geographical feature is captured using feeder taxonomy and hierarchical clustering. Furthermore, the accidental factor as a new parameter is introduced to evaluate the vandalism, vegetation, and operating fault to the grid. Finally, the warning information and advice for customers will be presented after fault chain analysis. The FP for a specific area in Australia is analyzed in case studies to verify the proposed model.