Risk Modeling Research Articles

Empirical modeling of agricultural climate risk

Effective policies for adaptation to climate change require understanding how impacts are related to exposures and vulnerability, the dimensions of the climate system that will change most and where human impacts will be most draconian, and the institutions best suited to respond. Here, we propose a simple method for more credibly pairing empirical statistical damage estimates derived from recent weather and outcome observations with projected future climate changes and proposed responses. We first analyze agricultural production and loan repayment data from Brazil to understand vulnerability to historical variation in the more predictable components of temperature and rainfall (trend and seasonality) as well as to shocks (both local and over larger spatial scales). This decomposed weather variation over the past two decades explains over 50% of the yield variation in major Brazilian crops and, critically, can be constructed in the same way for future climate projections. Combining our estimates with bias-corrected downscaled climate simulations for Brazil, we find increased variation in yields and revenues (including more bad years and worse outcomes) and higher agricultural loan default at midcentury. Results in this context point to two particularly acute dimensions of vulnerability: Intensified seasonality and local idiosyncratic shocks both contribute to worsening outcomes, along with a reduced capacity for spatially correlated ("covariate") shocks to ameliorate these effects through prices. These findings suggest that resilience strategies should focus on institutions such as water storage, financial services, and reinsurance.

Improving the accuracy of credit scoring models using an innovative Bayesian informative prior specification method

A new Bayesian informative prior specification method (BAF method – Bayesian priors using ARIMA forecasts) is proposed to introduce additional information into credit risk modelling and improve model predictive performance. We use logistic regression to model the probability of default of mortgage loans comparing the Bayesian approach with various priors and the frequentist approach. But unlike previous literature, we treat coefficient estimates in the probability of default model as stochastic time series variables. We build ARIMA models to forecast the coefficient values in future time periods and use these ARIMA forecasts as Bayesian informative priors. We find that the Bayesian models using this prior specification method produce more accurate predictions for the probability of default as compared to frequentist models and Bayesian models with other priors.

Factor Analysis Framework for Credit, Operational, and Market Risk Modeling

Factor Analysis Framework for Credit, Operational, and Market Risk Modeling

Developing and validating machine learning-based prediction models for frailty occurrence in those with chronic obstructive pulmonary disease.

Frailty is a medical syndrome caused by multiple factors, characterized by decreased strength, endurance, and diminished physiological function, resulting in increased susceptibility to dependence and/or death. Patients with chronic obstructive pulmonary disease (COPD) tend to be more vulnerable to frailty due to their physical and psychological burdens. Therefore, the aim of this study was to develop a reliable and accurate vulnerability risk prediction model for frailty in patients with COPD in order to improve the identification and prediction of patient frailty. The specific objectives of this study were to determine the prevalence of frailty in patients with COPD and develop a prediction model and evaluate its predictive power. Clinical information was analyzed using data from the 2018 China Health and Retirement Longitudinal Study (CHARLS) database, and 34 indicators, including behavioral factors, health status, mental health parameters, and various sociodemographic variables, were examined in the study. The adaptive synthetic sampling technique was used for unbalanced data. Three methods, ridge regressor, extreme gradient boosting (XGBoost) classifier, and random forest (RF) regressor, were used to filter predictors. Seven machine learning (ML) techniques including logistic regression (LR), support vector machines (SVM), multilayer perceptron, light gradient-boosting machine, XGBoost, RF, and K-nearest neighbors were used to analyze and determine the optimal model. For customized risk assessment, an online predictive risk modeling website was created, along with Shapley additive explanation (SHAP) interpretations. Depression, smoking, gender, social activities, dyslipidemia, asthma, and residence type (urban vs. rural) were predictors for the development of frailty in patients with COPD. In the test set, the XGBoost model had an area under the curve of 0.942 (95% confidence interval: 0.925-0.959), an accuracy of 0.915, a sensitivity of 0.873, and a specificity of 0.911, indicating that it was the best model. The ML predictive model developed in this study is a useful and easy-to-use instrument for assessing the vulnerability risk of patients with COPD and may aid clinical physicians in screening high-risk patients.

Where and how often do people touch train interiors? An investigation during the pandemic

The coronavirus disease-2019 pandemic highlighted infection and hygiene as risks in dense public spaces, including public transport vehicles. This study investigated passenger touching behaviour within metro vehicles to understand the fomite exposure of passengers. Using in-vehicle CCTV on three lines of the London Underground, this paper investigated the number of touches on different types of train interiors and investigated their relationship with passenger density and movement. In total, 1818 station sections and 16 891 passengers were observed cumulatively from May to December 2021. The results showed that on average for each type of interiors on Victoria and Jubilee lines, there were between 0.07 and 0.57 touches per passenger movement (i.e. boarding or alighting) for deep tube lines, whereas District line that uses larger carriages showed a different tendency. The results also suggested that for deep tube lines, the number of touches per person increased beyond a density of ∼1.5 standing passengers/m2 possibly because passengers may touch interiors more in order not to bump into other passengers. These findings can be used in infection risk modelling and can inform mitigation of infection risk of future pandemics.

The role of technology in supply chain risk management: Innovations and challenges in logistics

The review delves into the pivotal role of technology in revolutionizing supply chain risk management practices within the logistics sector. The paper explores the myriad of risks faced by modern supply chains, ranging from natural disasters and geopolitical tensions to cyber threats and disruptions in global trade patterns. It investigates how technological innovations such as blockchain, Internet of Things (IoT), artificial intelligence (AI), and predictive analytics are reshaping traditional risk management approaches by providing real-time visibility, data-driven insights, and proactive mitigation strategies. Through a comprehensive analysis, the study examines the transformative potential of these technologies in enhancing supply chain resilience, agility, and responsiveness to unforeseen disruptions. It highlights the benefits of leveraging blockchain technology for secure and transparent supply chain transactions, IoT sensors for real-time monitoring of goods in transit, AI algorithms for predictive risk modeling, and predictive analytics for identifying and mitigating potential disruptions before they escalate. Furthermore, the paper delves into the challenges and complexities associated with adopting and integrating these technologies into existing supply chain processes, including interoperability issues, data privacy concerns, and the need for specialized expertise. By providing practical insights, case studies, and best practices, the study aims to empower logistics professionals, policymakers, and industry stakeholders to harness the full potential of technology-driven solutions in managing supply chain risks effectively and ensuring business continuity in an increasingly uncertain and volatile global environment.

Assessment of Climate Change Risk on Insurance Portfolios

Purpose: The general purpose of this study was to investigate the assessment of climate change risk on insurance portfolios.
 Methodology: The study adopted a desktop research methodology. Desk research refers to secondary data or that which can be collected without fieldwork. Desk research is basically involved in collecting data from existing resources hence it is often considered a low cost technique as compared to field research, as the main cost is involved in executive’s time, telephone charges and directories. Thus, the study relied on already published studies, reports and statistics. This secondary data was easily accessed through the online journals and library.
 Findings: The findings reveal that there exists a contextual and methodological gap relating to climate change risk on insurance portfolios. The assessment of climate change risk on insurance portfolios reveals the significant challenges posed by increasingly frequent and severe climate-related events, impacting insurers' financial stability and policyholders' security. Key conclusions emphasize the imperative for insurers to enhance climate risk modeling, integrate climate considerations into decision-making processes, and foster collaboration with stakeholders to develop effective risk management solutions. Proactive measures are essential to address the systemic nature of climate risks, underscoring the urgency of collective action and innovation to build resilience and sustainability in insurance portfolios and broader society amidst a changing climate.
 Unique Contribution to Theory, Practice and Policy: The Prospect theory, Agency theory and the Resilience theory may be used to anchor future studies relating to climate change risk on insurance portfolios. Insurers are advised to invest in advanced risk modeling, diversify risk exposure across regions and lines of business, integrate climate risk into underwriting and pricing, strengthen regulatory oversight, collaborate with stakeholders, and invest in climate adaptation and mitigation measures. These recommendations aim to improve insurers' ability to quantify and manage climate-related risks effectively, promote sustainability, and contribute to broader resilience-building efforts in the face of climate change.
 Keywords: Climate Change Risk, Insurance Portfolios, Resilience, Risk Modeling, Diversification, Underwriting, Regulatory Oversight, Collaboration

Genetics of gambling disorder and related phenotypes: The potential uses of polygenic and multifactorial risk models to enable early detection and improve clinical outcomes.

Gambling Disorder (GD) is an impactful behavioural addiction for which there appear to be underpinning genetic contributors. Twin studies show significant GD heritability results and intergenerational transmission show high rates of transmission. Recent developments in polygenic and multifactorial risk prediction modelling provide promising opportunities to enable early identification and intervention for at risk individuals. People with GD often have significant delays in diagnosis and subsequent help-seeking that can compromise their recovery. In this paper we advocate for more research into the utility of polygenic and multifactorial risk modelling in GD research and treatment programs and rigorous evaluation of its costs and benefits.

Estimating crash risk and injury severity considering multiple traffic conflict and crash types: A bivariate extreme value approach

Traffic conflicts are generally considered independent events in existing extreme value theory models to estimate the risk of total or single types of crashes. However, traffic events at a road entity are not necessarily independent interactions and can lead to multiple traffic conflicts with shared common unobserved factors. A comprehensive estimation of crash risks in a road entity needs to consider the correlation of potential traffic conflicts to avoid possible bias in prediction performance and the problem of undetected deficiencies. This study proposes a Bayesian non-stationary bivariate generalised extreme value modelling framework to estimate the severe and non-severe crash risks accounting for the correlation between right-turn and rear-end conflicts at signalised intersections. A deep neural network-based computer vision technique was applied to extract the traffic conflicts from 77 h of video recordings over two right-turn approaches at two signalised intersections in Cairns, Australia. Post encroachment time and modified time to collision were used to characterise right-turn and rear-end conflicts, respectively, while an expected post-collision velocity difference was combined with post encroachment time and modified time to collision for crash risk estimation by injury severity levels. Several covariates were used to address the time-varying heterogeneity of traffic conflict extremes and to estimate the differential crash risks at signal cycles. Results showed a significant correlation between right-turn and rear-end crashes at signal cycle levels, indicating the importance of accounting for the dependency among traffic conflict types. Overall, the bivariate models considering the correlation among traffic conflict types were found to understandably perform better than their univariate counterparts. This study provides a demonstration of a correlated crash risk modelling framework that addresses issues related to the suitable traffic conflict measures, time varying risks (non-stationarity), heterogeneity, and injury severity levels of different crash types.

Agent-based risk modeling of domino effects in urban LNG stations

As the transfer hub between natural gas and users, urban LNG stations (ULSs) are the important link of the whole LNG industry chain. Due to special geographical location and various process equipment, potential risks in ULSs, especially domino effect accidents, usually lead to catastrophic consequences and social impacts. In this study, a dynamic risk assessment model for analyzing domino effects in urban LNG stations (DRADL) based on agent modelling and simulation is proposed. Tanks and tank cars as well as vaporizers and boilers in ULSs are modeled. A set of physical process and effect models are employed as the simulation process in this paper. Considering the lightweight nature and narrow flammability limit of LNG vapor cloud, a cyclic process is applied to judge the consequences of vapor cloud dispersion, which could avoid overestimating the results. A real urban LNG station is taken as a case study for modeling its potential domino effect risk. The results demonstrate the effectiveness and the advantages of the proposed ABMS approach for LNG dynamic risk assessment. The proposed DRADL model takes into account the combined effects of explosion and fire and LNG characteristics, which can provide guidance for risk management and rescue operation of domino effect accidents in urban LNG stations.

Introduction to the Special Track on Artificial Intelligence and COVID-19 (Abstract Reprint)

The human race is facing one of the most meaningful public health emergencies in the modern era caused by the COVID-19 pandemic. This pandemic introduced various challenges, from lock-downs with significant economic costs to fundamentally altering the way of life for many people around the world. The battle to understand and control the virus is still at its early stages yet meaningful insights have already been made. The uncertainty of why some patients are infected and experience severe symptoms, while others are infected but asymptomatic, and others are not infected at all, makes managing this pandemic very challenging. Furthermore, the development of treatments and vaccines relies on knowledge generated from an ever evolving and expanding information space. Given the availability of digital data in the modern era, artificial intelligence (AI) is a meaningful tool for addressing the various challenges introduced by this unexpected pandemic. Some of the challenges include: outbreak prediction, risk modeling including infection and symptom development, testing strategy optimization, drug development, treatment repurposing, vaccine development, and others.

THE ROLE OF TECHNOLOGY IN SUPPLY CHAIN RISK MANAGEMENT: INNOVATIONS AND CHALLENGES IN LOGISTICS

The review delves into the pivotal role of technology in revolutionizing supply chain risk management practices within the logistics sector. The paper explores the myriad of risks faced by modern supply chains, ranging from natural disasters and geopolitical tensions to cyber threats and disruptions in global trade patterns. It investigates how technological innovations such as blockchain, Internet of Things (IoT), artificial intelligence (AI), and predictive analytics are reshaping traditional risk management approaches by providing real-time visibility, data-driven insights, and proactive mitigation strategies. Through a comprehensive analysis, the study examines the transformative potential of these technologies in enhancing supply chain resilience, agility, and responsiveness to unforeseen disruptions. It highlights the benefits of leveraging blockchain technology for secure and transparent supply chain transactions, IoT sensors for real-time monitoring of goods in transit, AI algorithms for predictive risk modeling, and predictive analytics for identifying and mitigating potential disruptions before they escalate. Furthermore, the paper delves into the challenges and complexities associated with adopting and integrating these technologies into existing supply chain processes, including interoperability issues, data privacy concerns, and the need for specialized expertise. By providing practical insights, case studies, and best practices, the study aims to empower logistics professionals, policymakers, and industry stakeholders to harness the full potential of technology-driven solutions in managing supply chain risks effectively and ensuring business continuity in an increasingly uncertain and volatile global environment.
 Keywords: Technology, Supply Chain, Risk Management, Innovations, Challenges, Logistics.

Predicting the Length of Mechanical Ventilation in Acute Respiratory Disease Syndrome Using Machine Learning: The PIONEER Study.

Background: The ability to predict a long duration of mechanical ventilation (MV) by clinicians is very limited. We assessed the value of machine learning (ML) for early prediction of the duration of MV > 14 days in patients with moderate-to-severe acute respiratory distress syndrome (ARDS). Methods: This is a development, testing, and external validation study using data from 1173 patients on MV ≥ 3 days with moderate-to-severe ARDS. We first developed and tested prediction models in 920 ARDS patients using relevant features captured at the time of moderate/severe ARDS diagnosis, at 24 h and 72 h after diagnosis with logistic regression, and Multilayer Perceptron, Support Vector Machine, and Random Forest ML techniques. For external validation, we used an independent cohort of 253 patients on MV ≥ 3 days with moderate/severe ARDS. Results: A total of 441 patients (48%) from the derivation cohort (n = 920) and 100 patients (40%) from the validation cohort (n = 253) were mechanically ventilated for >14 days [median 14 days (IQR 8-25) vs. 13 days (IQR 7-21), respectively]. The best early prediction model was obtained with data collected at 72 h after moderate/severe ARDS diagnosis. Multilayer Perceptron risk modeling identified major prognostic factors for the duration of MV > 14 days, including PaO2/FiO2, PaCO2, pH, and positive end-expiratory pressure. Predictions of the duration of MV > 14 days showed modest discrimination [AUC 0.71 (95%CI 0.65-0.76)]. Conclusions: Prolonged MV duration in moderate/severe ARDS patients remains difficult to predict early even with ML techniques such as Multilayer Perceptron and using data at 72 h of diagnosis. More research is needed to identify markers for predicting the length of MV. This study was registered on 14 August 2023 at ClinicalTrials.gov (NCT NCT05993377).

Market risk modeling with option-implied covariances and score-driven dynamics

In this paper we make use of option-implied volatilities to build a time-varying implied correlation matrix. Then, we use this matrix to estimate jointly both the covariance matrix of the returns and the implied covariance matrix dynamics. Finally, we do a backtest and show that the proposed model can effectively use the risk-neutral information to model the variance of the returns and to forecast the Value-at-Risk. Our results show that, in general, the proposed model outperforms the benchmark while considerably reducing the number of parameters to be estimated.

CORR Insights®: Machine Learning Did Not Outperform Conventional Competing Risk Modeling to Predict Revision Arthroplasty.

CORR Insights®: Machine Learning Did Not Outperform Conventional Competing Risk Modeling to Predict Revision Arthroplasty.

Reliability risk modelling of asphalt pavement structure performance under the impact of freeze-thaw cycles

This paper delivers an in-depth risk analysis of asphalt pavement structural strength in the cold Qinghai-Tibet Plateau region. It encompasses a wide range of risk factors, including environmental conditions, traffic dynamics, pavement structures and materials, and various human-related factors. These risk factors are quantified using probability distributions, with emphasis on the changes in their mean and variance to simulate the effects of risk. A novel risk metric, termed “CVall”, is introduced to encapsulate the cumulative risk associated with each random variable. The paper describes the risk in the load and resistance equations, which govern the asphalt pavement strength, using the probability density function distribution. This approach takes into account the variation coefficient of each parameter. The study highlights that the risk related to the pavement structure number following freeze-thaw cycles significantly influences the reliability of pavement structural strength. These insights are particularly crucial for highway design, construction, and maintenance in cold regions, where the durability and longevity of pavement are paramount.

Unraveling trends in schistosomiasis: deep learning insights into national control programs in China.

To achieve the ambitious goal of eliminating schistosome infections, the Chinese government has implemented diverse control strategies. This study explored the progress of the 2 most recent national schistosomiasis control programs in an endemic area along the Yangtze River in China. We obtained village-level parasitological data from cross-sectional surveys combined with environmental data in Anhui Province, China from 1997 to 2015. A convolutional neural network (CNN) based on a hierarchical integro-difference equation (IDE) framework (i.e., CNN-IDE) was used to model spatio-temporal variations in schistosomiasis. Two traditional models were also constructed for comparison with 2 evaluation indicators: the mean-squared prediction error (MSPE) and continuous ranked probability score (CRPS). The CNN-IDE model was the optimal model, with the lowest overall average MSPE of 0.04 and the CRPS of 0.19. From 1997 to 2011, the prevalence exhibited a notable trend: it increased steadily until peaking at 1.6 per 1,000 in 2005, then gradually declined, stabilizing at a lower rate of approximately 0.6 per 1,000 in 2006, and approaching zero by 2011. During this period, noticeable geographic disparities in schistosomiasis prevalence were observed; high-risk areas were initially dispersed, followed by contraction. Predictions for the period 2012 to 2015 demonstrated a consistent and uniform decrease. The proposed CNN-IDE model captured the intricate and evolving dynamics of schistosomiasis prevalence, offering a promising alternative for future risk modeling of the disease. The comprehensive strategy is expected to help diminish schistosomiasis infection, emphasizing the necessity to continue implementing this strategy.

Machine Learning Did Not Outperform Conventional Competing Risk Modeling to Predict Revision Arthroplasty.

Estimating the risk of revision after arthroplasty could inform patient and surgeon decision-making. However, there is a lack of well-performing prediction models assisting in this task, which may be due to current conventional modeling approaches such as traditional survivorship estimators (such as Kaplan-Meier) or competing risk estimators. Recent advances in machine learning survival analysis might improve decision support tools in this setting. Therefore, this study aimed to assess the performance of machine learning compared with that of conventional modeling to predict revision after arthroplasty. Does machine learning perform better than traditional regression models for estimating the risk of revision for patients undergoing hip or knee arthroplasty? Eleven datasets from published studies from the Dutch Arthroplasty Register reporting on factors associated with revision or survival after partial or total knee and hip arthroplasty between 2018 and 2022 were included in our study. The 11 datasets were observational registry studies, with a sample size ranging from 3038 to 218,214 procedures. We developed a set of time-to-event models for each dataset, leading to 11 comparisons. A set of predictors (factors associated with revision surgery) was identified based on the variables that were selected in the included studies. We assessed the predictive performance of two state-of-the-art statistical time-to-event models for 1-, 2-, and 3-year follow-up: a Fine and Gray model (which models the cumulative incidence of revision) and a cause-specific Cox model (which models the hazard of revision). These were compared with a machine-learning approach (a random survival forest model, which is a decision tree-based machine-learning algorithm for time-to-event analysis). Performance was assessed according to discriminative ability (time-dependent area under the receiver operating curve), calibration (slope and intercept), and overall prediction error (scaled Brier score). Discrimination, known as the area under the receiver operating characteristic curve, measures the model's ability to distinguish patients who achieved the outcomes from those who did not and ranges from 0.5 to 1.0, with 1.0 indicating the highest discrimination score and 0.50 the lowest. Calibration plots the predicted versus the observed probabilities; a perfect plot has an intercept of 0 and a slope of 1. The Brier score calculates a composite of discrimination and calibration, with 0 indicating perfect prediction and 1 the poorest. A scaled version of the Brier score, 1 - (model Brier score/null model Brier score), can be interpreted as the amount of overall prediction error. Using machine learning survivorship analysis, we found no differences between the competing risks estimator and traditional regression models for patients undergoing arthroplasty in terms of discriminative ability (patients who received a revision compared with those who did not). We found no consistent differences between the validated performance (time-dependent area under the receiver operating characteristic curve) of different modeling approaches because these values ranged between -0.04 and 0.03 across the 11 datasets (the time-dependent area under the receiver operating characteristic curve of the models across 11 datasets ranged between 0.52 to 0.68). In addition, the calibration metrics and scaled Brier scores produced comparable estimates, showing no advantage of machine learning over traditional regression models. Machine learning did not outperform traditional regression models. Neither machine learning modeling nor traditional regression methods were sufficiently accurate in order to offer prognostic information when predicting revision arthroplasty. The benefit of these modeling approaches may be limited in this context.

Are Femoral Stems in Primary Total Knee Arthroplasty Cost Effective in High Fracture Risk Patients? A Risk Model and Cost Analysis.

Femoral stemmed total knee arthroplasty (FS TKA) may be used in patients deemed higher risk for periprosthetic fracture (PPF) to reduce PPF risk. However, the cost effectiveness of FS TKA has not been defined. Using a risk modeling analysis, we investigate the cost effectiveness of FS in primary TKA compared with the implant cost of revision to distal femoral replacement (DFR) following PPF. A model of risk categories was created representing patients at increasing fracture risk, ranging from 2.5 to 30%. The number needed to treat (NNT) was calculated for each risk category, which was multiplied by the increased cost of FS TKA and compared with the cost of DFR. The 50th percentile implant pricing data for primary TKA, FS TKA, and DFR were identified and used for the analysis. FS TKA resulted in an increased cost of $2,717.83, compared with the increased implant cost of DFR of $27,222.29. At 50% relative risk reduction with FS TKA, the NNT for risk categories of 2.5, 10, 20, and 30% were 80, 20, 10, and 6.67, respectively. At 20% risk, FS TKA times NNT equaled $27,178.30. A 10% absolute risk reduction in fracture risk obtained with FS TKA is needed to achieve cost neutrality with DFR. FS TKA is not cost effective for low fracture risk patients but may be cost effective for patients with fracture risk more than 20%. Further study is needed to better define the quantifiable risk reduction achieved in using FS TKA and identify high-risk PPF patients.

An open-source radar-based hail damage model for buildings and cars

Abstract. Severe hailstorms result in substantial damage to buildings and vehicles, necessitating the quantification of associated risks. Here, we present a novel open-source hail damage model for buildings and cars based on single-polarization radar data and 250 000 geolocated hail damage reports in Switzerland from 2002 to 2021. To this end, we conduct a detailed evaluation of different radar-based hail intensity measures at 1 km resolution and find that the maximum expected severe hail size (MESHS) outperforms the other measures, despite a considerable false-alarm ratio. Asset-specific hail damage impact functions for buildings and cars are calibrated based on MESHS and incorporated into the open-source risk modelling platform CLIMADA. The model successfully estimates the correct order of magnitude for the number of damaged building in 91 %, their total cost in 77 %, the number of damaged vehicles in 74 %, and their total cost in 60 % of over 100 considered large hail events. We found considerable uncertainties in hail damage estimates, which are largely attributable to limitations of radar-based hail detection. Therefore, we explore the usage of crowdsourced hail reports and find substantially improved spatial representation of severe hail for individual events. By highlighting the potential and limitations of radar-based hail size estimates, particularly MESHS, and the utilization of an open-source risk modelling platform, this study represents a significant step towards addressing the gap in risk quantification associated with severe hail events in Switzerland.