Article
Computer Science, Artificial Intelligence
Ricardo Munoz-Cancino, Cristian Bravo, Sebastian A. Rios, Manuel Grana
Summary: Credit risk management has been using credit scoring models at different stages for over half a century. Social network data has been shown to increase the predictive power of these models, especially when historical data is limited. This study analyzes the dynamics of creditworthiness assessment and finds that credit scoring based on borrowers' history improves performance initially and then stabilizes. The use of social network features adds value to credit scoring for loan applications and throughout the study period for business scoring.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Economics
Paul A. Grout
Summary: The adoption of AI and ML in financial markets may not necessarily increase competition, and is unlikely to lead to significant changes in concentration in the short term. Whether there will be more concentration in the longer term depends on the balance of two opposing forces, while the impact of adopting AI and ML at the micro-market level is highly sensitive to the specifics of the market.
OXFORD REVIEW OF ECONOMIC POLICY
(2021)
Article
Management
Elena Dumitrescu, Sullivan Hue, Christophe Hurlin, Sessi Tokpavi
Summary: In the context of credit scoring, a high-performance and interpretable credit scoring method called penalised logistic tree regression (PLTR) is proposed, which uses information from decision trees to improve logistic regression performance, allowing for capturing non-linear effects in credit scoring data while maintaining interpretability.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2022)
Article
Business, Finance
Katerina Rigana, Ernst-Jan Camiel Wit, Samantha Cook
Summary: Contagion plays a crucial role in finance, as it is the core of most major financial crises, especially the global financial crisis in 2007. This study introduces a new measure for quantifying contagion among individual currencies in the Foreign exchange market and demonstrates the causal pathways of contagion using causal inference. By identifying the sources of contagion and assessing the diversification options and susceptibility to systemic risk of different currencies, this research provides insights into the level of global systemic risk, with a particular focus on the impacts of the Covid-19 pandemic.
INTERNATIONAL REVIEW OF FINANCIAL ANALYSIS
(2023)
Article
Business, Finance
Boqiang Lin, Rui Bai
Summary: This paper investigates the relationship between financial and non-financial indicators of 40 listed enterprises in the mining, steel, and power industries and debt financing. Using the XGBoost method, the top six indicators are identified for predicting long-term debt. Further explanation is provided based on the Shapley additive explanation value.
FINANCE RESEARCH LETTERS
(2022)
Review
Computer Science, Information Systems
Asma Cherif, Arwa Badhib, Heyfa Ammar, Suhair Alshehri, Manal Kalkatawi, Abdessamad Imine
Summary: Credit card fraud is a serious problem due to emerging technologies like contactless payment. This article provides an in-depth review of cutting-edge research from 2015 to 2021 on detecting and predicting fraudulent credit card transactions. The study reveals limited investigation into deep learning, highlighting the need for more research to address challenges in detecting fraud using new technologies. This study serves as a valuable resource for academic and industrial researchers in evaluating fraud detection systems and designing solutions.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2023)
Article
Management
Michael Buecker, Gero Szepannek, Alicja Gosiewska, Przemyslaw Biecek
Summary: Credit scoring models require accurate risk prediction and transparency, but the superior predictive power of modern machine learning algorithms is not fully utilized. This article presents a framework for making black box machine learning models transparent, auditable, and explainable, and shows how these techniques can be applied in credit scoring to achieve comparable interpretability while maintaining predictive power.
JOURNAL OF THE OPERATIONAL RESEARCH SOCIETY
(2022)
Editorial Material
Food Science & Technology
Yuanchao Fan
Summary: Model simulations indicate regional disparities in crop production responses to solar geoengineering scenarios that use sunlight reflection to cool the Earth.
Article
Computer Science, Artificial Intelligence
Isaac Tonkin, Adrian Gepp, Geoff Harris, Bruce Vanstone
Summary: This paper extends deep learning models developed on the US equity data to the Australian market and finds that these models are relatively less accurate at predicting next day returns compared to the original models. By modifying and training the models on Australian data, the paper identifies the best-performing models and attributes the improvement to regional influences within the training data sets. This finding suggests the importance of considering market-specific bias in developing deep learning models.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Chemistry, Multidisciplinary
Jacobo Chaquet-Ulldemolins, Francisco-Javier Gimeno-Blanes, Santiago Moral-Rubio, Sergio Munoz-Romero, Jose-Luis Rojo-Alvarez
Summary: Artificial intelligence has become increasingly important in the global economy, especially in the field of credit fraud detection. Autoencoders have proven to be effective in discovering nonlinear features, but are often seen as black boxes. This study proposes an interpretable and agnostic methodology for credit fraud detection, using a novel technique based on autoencoders. Results show improved accuracy compared to previous models, and the methodology allows for individualized and unbiased analysis.
APPLIED SCIENCES-BASEL
(2022)
Article
Business, Finance
Alanoud Al-Maadid, Saleh Alhazbi, Khaled Al-Thelaya
Summary: This study investigates the impact of COVID-19-related news on stock markets in GCC countries using machine learning techniques. The results show that the stock markets in UAE, Qatar, Saudi Arabia, and Oman were influenced by coronavirus news, while no impact was observed in Bahrain. Furthermore, the affected markets were influenced differently in terms of the quantities and types of news.
RESEARCH IN INTERNATIONAL BUSINESS AND FINANCE
(2022)
Article
Economics
Paulo Vitor de Campos Souza, Luiz Carlos Bambirra Torres
Summary: Extreme Learning Machines enable multilayer neural networks to improve efficiency, particularly crucial in the financial sector for identifying defaulters in credit card financial relationships.
COMPUTATIONAL ECONOMICS
(2021)
Article
Computer Science, Artificial Intelligence
Cheng-Feng Wu, Shian-Chang Huang, Chei-Chang Chiou, Yu-Min Wang
Summary: The study demonstrates that the deep multiple kernel classifier outperforms conventional and ensemble models in credit risk assessment, helping credit card issuers appropriately approve applicants for credit cards, improving risk management, avoiding bad debt, and benefiting banks.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Artificial Intelligence
Furkan Baser, Oguz Koc, Sevtap Selcuk-Kestel
Summary: This study proposes a clustering based fuzzy classification (CBFC) method for credit risk assessment, which enhances the prediction power of machine learning methods by adopting fuzzy theory. Extensive comparisons are performed to demonstrate the performance of CBFC compared to traditional methods on datasets with different characteristics. The findings show that CBFC models can produce promising results in credit risk evaluation, aiding practitioners and decision makers in credit issuance.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Management
Panagiotis Avramidis, Nikolaos Mylonopoulos, George G. Pennacchi
Summary: This paper develops a model to study the competition between banks and marketplace lenders using local market data in the U.S. Banks and the largest marketplace lending platform are analyzed. The results show that marketplace lending can absorb unmet consumer credit demand when bank credit availability declines. It is also found that marketplace lending helps mitigate credit distress caused by bank mergers in local economies.
MANAGEMENT SCIENCE
(2022)