Default Model Research Articles

Alteration of segregation of brain systems in the severe depressive disorder after electroconvulsive therapy

• The investigation assessed the changes in FNC of internetwork low-level cognitive networks associated with ECT response in SDD, and further analyzed the relation between advanced cognitive networks and low-level cognitive network. • ECT was able to improve abnormal interactions between frontal-parietal network (FPN) and other function systems in participants with SDD. • Compared with the healthy people after ECT, the connection between the advanced cognitive network and low-level cognitive network were changed, as well as that among low-level cognitive networks. • Functional connectivity of within DMN and FPN systems were strengthened in after-ECT SDD compared to before-ECT SDD. Severe depressive disorder (SDD) has been characterized by abnormal brain activity and interactions across the whole-brain functional networks. Electroconvulsive therapy (ECT) has been considered as an effective and rapid treatment for SDD. However, whether ECT may enhance the dysconnectivity between brain networks to achieve antidepressant effects is still largely unknown. Resting-state functional magnetic resonance imaging was acquired from 25 SDD participants (9M/16F, age: 34.32±9.69) before and after ECT treatments (9M/16F, age: 34.32±9.69), along with 25 sex-and age-matched healthy controls (9M/16F, age: 24.72±8.16). Then, large-scale functional brain network of each participant was constructed and divided into ten functional systems. Finally, the topology properties across ten systems, including with- and between-system connectivity, system-segregation and participation coefficient, were assessed and analyzed between SDD patients and controls. Our results showed ECT can enhance the reduced functional connectivity (FC) of default model network (DMN) and frontal-parietal network (FPN) in SDD patients. In particular, after ECT, not only the FC within DMN and FPN was strengthened, but also the FC between DMN/FPN with subcortical network (SUB) and visual network was significantly increased. These above connectives of DMN and FPN were significantly reduced in before-ECT SDD compared to healthy controls and were closely related to clinical scores of depression. Our findings indicated the enhanced interactions in with- and between-systems of DMN and FPN may contributed to the ECT response in SDD patients. And These findings provide novel and important insights into the neurobiological mechanisms underlying depression, even ECT.

Quantification of model risk with an application to probability of default estimation and stress testing for a large corporate portfolio

This paper addresses the building of obligor-level hazard rate corporate probability of default models for stress testing, departing from the predominant practice in wholesale credit modeling of constructing segment-level models for this purpose. We build models based upon varied financial, credit rating, equity market and macroeconomic factors, using an extensive history of large corporate firms sourced from Moody’s.We develop distance-to-default (DTD) risk factors and design hybrid structural/ Merton reduced-form models as challengers to versions of the models containing other individual variables. We measure the model risk attributed to various modeling assumptions according to the principle of relative entropy and observe that the omitted-variable bias with respect to the DTD risk factor, neglect of interaction effects and incorrect link function specification has the greatest, intermediate and least impact, respectively. Given the sensitive regulatory uses of these models and the concerns raised in the industry about the effect of model misspecification on capital and reserves, our conclusion is that validation methods chosen in the stress testing context should be capable of testing model assumptions. Our research adds to the literature in that it offers state-of-the-art techniques as viable options in the arsenal of model validators, developers and supervisors seeking to manage model risk.

A general firm value model under partial information

We introduce a new structural default model intended to combine enhanced economic relevance and affordable computational complexity. Our approach exploits the information conveyed by a noisy observation of the firm’s value combined with the firm’s actual default state. Moreover, the model is reasonably general, since any diffusion can be used to depict the firm’s dynamics. However, this realistic setup comes at the cost of significant computational challenges. To mitigate these, we propose an implementation based on recursive quantization. A thorough analysis of the approximation error resulting from our numerical procedure is provided.We find that the observation noise has a significant impact on the credit spreads’ implied volatilities. The power of our method is illustrated through the pricing of credit default swap options.

A Unified Framework for Default Modeling

A Unified Framework for Default Modeling

The Bribe Rate and Long Run Differences in Sovereign Borrowing Costs

The average cost of borrowing on international financial markets varies widely from nation to nation even after controlling for the varying levels of indebtedness of their governments. This suggests that markets assign country specific default risk assessments. In this paper we focus on one natural source of this difference – the quality of their institutions. We begin by showing that the average sovereign spread is positively related to the average percentage of a government contract that must be given as a “gift” in order to secure the contract. We then build a sovereign default model where the government is constrained to use corrupt bureaucrats to deliver public goods and services and manage its accounts. Using the gift data as a measure of the public resources diverted by bureaucrats, and the Rule of Law index as a measure of institutional quality, we estimate the diversion policy of bureaucrats and use it to calibrate our model to international data. We use the model to generate an artificial international data set where countries vary only in the institutional quality parameter. Running the same regressions on our artificial data, we find that average spreads are positively associated with the bribe (gift) rate and with average debt levels. The model implies that when revenue streams are low, a benefit of default is that public services net of bureaucratic diversion actually increase. Since the net gain in public services obtained by defaulting is decreasing in the level of institutional quality, international lenders assign lower default risk to those countries.

Assume a Chimpanzee Chimpanzee Behavior is Consistent with the Default Model of Economic Choice, But Human Behavior is Not

Assume a Chimpanzee Chimpanzee Behavior is Consistent with the Default Model of Economic Choice, But Human Behavior is Not

Comparing the Effect of Intangible Assets and Macroeconomic Parameters on the Stock Prices in Firms Listed on the Stock Exchange of Iran, Saudi Arabia and Iraq

Comparing the Effect of Intangible Assets and Macroeconomic Parameters on the Stock Prices in Firms Listed on the Stock Exchange of Iran, Saudi Arabia and Iraq

Validation of corporate probability of default models considering alternative use cases and the quantification of model risk

<abstract> <p>In this study we consider the construction of <italic>through-the-cycle</italic> ("TTC") <italic>probability-of-default</italic> ("PD") models designed for credit underwriting uses and <italic>point-in-time</italic> ("PIT") PD models suitable for early warning uses, considering which validation elements should be emphasized in each case. We build PD models using a long history of large corporate firms sourced from Moody's, with a large number of financial, equity market and macroeconomic candidate explanatory variables. We construct a Merton model-style <italic>distance-to-default</italic> ("DTD") measure and build hybrid structural and reduced-form models to compare with the financial ratio and macroeconomic variable-only models. In the hybrid models, the financial and macroeconomic explanatory variables still enter significantly and improve the predictive accuracy of the TTC models, which generally lag behind the PIT models in that performance measure. While all classes of models have high discriminatory power by most measures, on an out-of-sample basis the TTC models perform better than the PIT models. We measure the model risk attributable to various model assumptions according to the principle of <italic>relative entropy</italic> and observe that omitted variable bias with respect to the DTD risk factor, neglect of interaction effects and incorrect link function specification has the greatest, intermediate and least impacts, respectively. We conclude that care must be taken to judiciously choose how we validate TTC vs. PIT models, as criteria may be rather different and apart from standards such as discriminatory power. This study contributes to the literature by providing expert guidance to credit risk modeling, model validation and supervisory practitioners in managing model risk.</p> </abstract>

The problem of psychological help to a person in the face of modern challenges

The problem of psychological help to a person in the context of modern challenges such as the COVID-19 pandemic, is considered from the point of view of compliance of this help's content to the public demand for it. The article aims to substantiate the need for a specific revision of traditional targets and traditional practices of psychological care due to the challenges that arise in the modern world. Reconstruction of the public request for psychological help is carried out according to the mass surveys conducted under the author's leadership: a nationwide representative study, which covered 1202 people aged from 18 y.o. and older, and an online survey conducted directly during the spring lockdown 2020, which involved 565 respondents. The analysis also includes data from other studies conducted during this period and following its "hot tracks." The article emphasizes that Ukrainian psychologists have shown significant professional activity since the corona crisis. Still, the effectiveness of their efforts was reduced by the fact that these efforts were exposed to barriers of both objective and subjective terms. In the psychological community, the default model of a person in a lockdown is focused on the psychological help implementation in the traditional content field of the individual (personality-oriented), marital, and family counseling. This model determined the main trend of psychologists' counseling activity in media and social networks, but, according to the obtained data, it did not fully meet the realities of life of Ukrainian citizens, the challenges of the information environment in which they were, and therefore their real needs. It is argued that global cataclysms such as the COVID-19 pandemic radically change the nature and principles of the psychologist's practical work, as they sharply shift the focus of their counseling activity from the usual axial to the retial communication systems. It is shown that the semantic field of psychological counseling in the conditions of modern challenges should be supplemented by at least two components – media-psychological and psychological-organizational. The need to launch a targeted state program to monitor mass sentiment and the psycho-emotional state of society is substantiated. The study's practical significance is to use its results and conclusions to improve psychological assistance to the population. Prospects for further development of the problem are seen in the definition of supervisory and technological prerequisites for expanding the content field of psychological counseling.

The Success of Managing Small Project in Light of Achieving the Dimensions of Managerial Leadership (Study Analysis of a Sample of Decision Makers’ Views in the General Company for Food Industries)

The aim of the research is to identify the role of administrative leadership in achieving the dimensions of the success of small projects’ management in the general company of foodstuff industries. The various variables have shown the importance of managerial leadership to achieve the dimensions of the success of the management of small projects, whether in international or local companies. The role of managerial leadership in the success of the management of small projects "In order to answer this question, has been developed a default model reflects the relationship and impact between management leadership and the success of the management of small projects. The research was based on descriptive analytical approach. The sample of the study consists of (48) Managers of departments and divisions where the research reached a number of results; the most important one is the existence of a relationship between the leadership of innovation and the success of the management of small projects in the company investigated if the relationship is positive and strong. The results led to a set of recommendations, the most important one is the necessity for the company under study to get benefit from the evaluation of the company’s leaders from time to time in order to achieve the dimensions of the success of managing small projects.

Evaluation of Biogenic Organic Aerosols in the Amazon Rainforest Using WRF‐Chem With MOSAIC

AbstractBiogenic secondary organic aerosol (BSOA) accounts for 70%–80% of submicron aerosol over the Amazon rainforest. However, uncertainties with regard to the chemical formation pathway, mass budget, and radiative impact of BSOA in the region still remain high. To address the issue, we used a regional chemistry transport model (WRF‐Chem‐MOSAIC) to evaluate the budget of BSOA in the Amazon forest during the wet season (March) of 2014 when the air mass was clean from influences from anthropogenic and biomass burning emissions. We found that the default model underestimated the surface concentration of submicron organic aerosol (OA) by 57% at a ground measurement site (T3). We subsequently updated the WRF‐Chem‐MOSAIC model by including SOA formation from isoprene epoxydiols (IEPOX‐SOA) and SOA aging schemes. The updated model reproduced the observed submicron OA concentration well with an overestimation of only 2.5%. With the updated model, we will show that IEPOX‐SOA accounts for 23% of surface submicron OA averaged over the entire Amazon rainforest. The modeled surface SOA (smaller than 2.5 μm in diameter) concentration in the Amazon region was about a factor of 14 larger than the global mean. Finally, a sensitivity study was conducted which showed that convective plumes can remove more than 40% of submicron OA below 5 km in the Amazon region. Overall, our study highlights the importance of IEPOX chemistry as well as the convective removal processes in the global SOA budget.

Investigating Explanatory Factors of Machine Learning Models for Plant Classification.

Recent progress in machine learning and deep learning has enabled the implementation of plant and crop detection using systematic inspection of the leaf shapes and other morphological characters for identification systems for precision farming. However, the models used for this approach tend to become black-box models, in the sense that it is difficult to trace characters that are the base for the classification. The interpretability is therefore limited and the explanatory factors may not be based on reasonable visible characters. We investigate the explanatory factors of recent machine learning and deep learning models for plant classification tasks. Based on a Daucus carota and a Beta vulgaris image data set, we implement plant classification models and compare those models by their predictive performance as well as explainability. For comparison we implemented a feed forward convolutional neuronal network as a default model. To evaluate the performance, we trained an unsupervised Bayesian Gaussian process latent variable model as well as a convolutional autoencoder for feature extraction and rely on a support vector machine for classification. The explanatory factors of all models were extracted and analyzed. The experiments show, that feed forward convolutional neuronal networks ( and mean accuracy) outperforms the Bayesian Gaussian process latent variable pipeline ( and mean accuracy) as well as the convolutional autoenceoder pipeline ( and mean accuracy) based approaches in terms of classification accuracy, even though not significant for Beta vulgaris images. Additionally, we found that the neuronal network used biological uninterpretable image regions for the plant classification task. In contrast to that, the unsupervised learning models rely on explainable visual characters. We conclude that supervised convolutional neuronal networks must be used carefully to ensure biological interpretability. We recommend unsupervised machine learning, careful feature investigation, and statistical feature analysis for biological applications.

Cosmic-ray positrons strongly constrain leptophilic dark matter

Cosmic-ray positrons have long been considered a powerful probe of dark matter annihilation. In particular, myriad studies of the unexpected rise in the positron fraction have debated its dark matter or pulsar origins. In this paper, we instead examine the potential for extremely precise positron measurements by AMS-02 to probe hard leptophilic dark matter candidates that do not have spectral features similar to the bulk of the observed positron excess. Utilizing a detailed cosmic-ray propagation model that includes a primary positron flux generated by Galactic pulsars in addition to a secondary component constrained by He and proton measurements, we produce a robust fit to the local positron flux and spectrum. We find no evidence for a spectral bump correlated with leptophilic dark matter, and set strong constraints on the dark matter annihilation cross-section that fall below the thermal annihilation cross-section for dark matter masses below 60 GeV and 380 GeV for annihilation into τ+τ- and e+e-, respectively, in our default model.

A methodological approach to developing and validating IFRS 9 -LGD parameters

Abstract Since the introduction of the advanced internal rating based approach through the Basel framework, financial institutions and regulators have been dealing with the increased complexity of Loss Given Default models. The development and validation of the parameters has become more formalized and standardized as more prescriptive regulations and guidelines have been published by the European Parliament, European Central Bank and European Banking Authority. Furthermore, following the introduction of IFRS 9 in January 2018 even more emphasis is put on the development and validation as the standard poses new challenges to the way models are designed, developed, validated and increased complexity through the introduction of the lifetime and forward-looking adjustments. This paper address the challenges faced by banks and regulators when assessing and validation the IFRS 9 - Loss Given Default parameters and framework. Moreover, it describes a non-exhaustive list of tests that can be performed to establishing the accuracy, discrimination power and stability of the Loss Given Default outputs. It is important that the framework built around the modeling, development and validation process allows models to be easily integrated in the management framework in a flexible manner that can facilitate any changes that must be brought to the models. Hence, this paper outlines a non-exhaustive list of quantitative validation tests considered would satisfy the requirements of the IFRS 9 standard.

Validation of Corporate Probability of Default Models Considering Alternative Use Cases

In this study, we consider the construction of through-the-cycle (“TTC”) PD models designed for credit underwriting uses and point-in-time (“PIT”) PD models suitable for early warning uses, considering which validation elements should be emphasized in each case. We build PD models using a long history of large corporate firms sourced from Moody’s, with a large number of financial, equity market and macroeconomic variables as candidate explanatory variables. We construct a Merton model-style distance-to-default (“DTD”) measure and build hybrid structural reduced-form models to compare with the financial ratio and macroeconomic variable-only models. In the hybrid models, the financial and macroeconomic explanatory variables still enter significantly and improve the predictive accuracy of the TTC models, which generally lag behind the PIT models in that performance measure. We conclude that care must be taken to judiciously choose the manner in which we validate TTC vs. PIT models, as criteria may be rather different and be apart from standards such as discriminatory power. This study contributes to the literature by providing expert guidance to credit risk modeling, model validation and supervisory practitioners in controlling the model risk associated with such modeling efforts.

Modelling of long-term retention of high-fired plutonium oxide in the human respiratory tract: importance of scar-tissue compartments

The U.S. Transuranium and Uranium Registries whole-body tissue donor Case 0407 had an acute intake of ‘high-fired’ plutonium oxide resulting from a glove-box fire in a fabrication plant at a nuclear defence facility. The respiratory tract of this individual was dissected into five regions (larynx, bronchi, bronchioles, alveolar-interstitial, and thoracic lymph nodes) and analysed for plutonium content. The activities in certain compartments of the respiratory tract were found to be higher than expected from the default models described in publications of the International Commission on Radiological Protection. Because of the extremely slow rate of dissolution of the material inhaled, the presence of bound fraction is incapable of explaining the higher-than-expected retention. A plausible hypothesis—encapsulation of plutonium in scar tissues—is supported by the review of literature. Therefore, scar-tissue compartments corresponding to the larynx, bronchi, bronchioles and alveolar-interstitial regions were added to the existing human respiratory tract model structure. The transfer rates between these compartments were determined using Markov Chain Monte Carlo analysis of data on urinary excretion, lung counts and post-mortem measurements of the liver, skeleton and regional retention in the respiratory tract. Modelling of the data showed that approximately 30% of plutonium activity in the lung was sequestered in scar tissues. The dose consequence of such sequestration is qualitatively compared against that of chemical binding.

Development of an Impairment Point in Time Probability of Default Model for Revolving Retail Credit Products: South African Case Study

A new methodology to derive IFRS 9 PiT PDs is proposed. The methodology first derives a PiT term structure with accompanying segmented term structures. Secondly, the calibration of credit scores using the Lorenz curve approach is used to create account-specific PD term structures. The PiT term structures are derived by using empirical information based on the most recent default information and account risk characteristics prior to default. Different PiT PD term structures are developed to capture the structurally different default risk patterns for different pools of accounts using segmentation. To quantify what a materially different term structure constitutes, three tests are proposed. Account specific PiT PDs are derived through the Lorenz curve calibration using the latest default experience and credit scores. The proposed methodology is illustrated on an actual dataset, using a revolving retail credit portfolio from a South African bank. The main advantages of the proposed methodology include the use of well-understood methods (e.g., Lorenz curve calibration, scorecards, term structure modelling) in the banking industry. Further, the inclusion of re-default events in the proposed IFRS 9 PD methodology will simplify the development of the accompanying IFRS 9 LGD model due to the reduced complexity for the modelling of cure cases. Moreover, attrition effects are naturally included in the PD term structures and no longer require a separate model. Lastly, the PD term structure is based on months since observation, and therefore the arrears cycle could be investigated as a possible segmentation.

The Pandemic and Shareholder Value

Shareholder value has driven corporate governance in North America for over a century. In the wake of significant financial crises and growing inequalities, corporate America decided in 2019 to embrace a more egalitarian model, in which all stakeholders matter equally. The brutal pandemic that wreaked havoc in the first half of 2020 exposed a startling disconnect between the real economy and the stock market. This disconnect is due to a gap between explicit and implicit corporate governance. While officially corporate America wants to convert to a new doctrine, the pandemic has shown that shareholder capitalism has remained the default model. Good intentions and official declarations are not enough in a system that has been specifically designed to serve the shareholders. If stakeholder capitalism is to succeed, it needs a clear normative content and perhaps a more radical reform of institutions and regulation.

An Improved Deep Convolutional Neural Network-Based Autonomous Road Inspection Scheme Using Unmanned Aerial Vehicles

Recent advancements in the field of machine learning (ML) provide opportunity to conduct research on autonomous devices for a variety of applications. Intelligent decision-making is a critical task for self-driving systems. An attempt is made in this study to use a deep learning (DL) approach for the early detection of road cracks, potholes, and the yellow lane. The accuracy is not sufficient after training with the default model. To enhance accuracy, a convolutional neural network (CNN) model with 13 convolutional layers, a softmax layer as an output layer, and two fully connected layers (FCN) are constructed. In order to achieve the deeper propagation and to prevent saturation in the training phase, mish activation is employed in the first 12 layers with a rectified linear unit (ReLU) activation function. The upgraded CNN model performs better than the default CNN model in terms of accuracy. For the varied situation, a revised and enriched dataset for road cracks, potholes, and the yellow lane is created. The yellow lane is detected and tracked in order to move the unmanned aerial vehicle (UAV) autonomously by following yellow lane. After identifying a yellow lane, the UAV performs autonomous navigation while concurrently detecting road cracks and potholes using the robot operating system within the UAV. The performance model is benchmarked using performance measures, such as accuracy, sensitivity, F1-score, F2-score, and dice-coefficient, which demonstrate that the suggested technique produces better outcomes.

Using Model Performance to Assess the Representativeness of Data for Model Development and Calibration in Financial Institutions

This paper proposes a methodology that utilises model performance as a metric to assess the representativeness of external or pooled data when it is used by banks in regulatory model development and calibration. There is currently no formal methodology to assess representativeness. The paper provides a review of existing regulatory literature on the requirements of assessing representativeness and emphasises that both qualitative and quantitative aspects need to be considered. We present a novel methodology and apply it to two case studies. We compared our methodology with the Multivariate Prediction Accuracy Index. The first case study investigates whether a pooled data source from Global Credit Data (GCD) is representative when considering the enrichment of internal data with pooled data in the development of a regulatory loss given default (LGD) model. The second case study differs from the first by illustrating which other countries in the pooled data set could be representative when enriching internal data during the development of a LGD model. Using these case studies as examples, our proposed methodology provides users with a generalised framework to identify subsets of the external data that are representative of their Country’s or bank’s data, making the results general and universally applicable.