scholarly journals Credit Risk Modeling: Combining Classification And Regression Algorithms to Predict Expected Loss

2014 ◽  
Vol 8 (4) ◽  
pp. 4-10
Author(s):  
Tim Kreienkamp ◽  
Andrey Kateshov
Keyword(s):  
Credit Risk ◽  
Risk Model ◽  
Machine Learning Techniques ◽  
Risk Modeling ◽  
Expected Loss ◽  
Financial Industry ◽  
Loan Default ◽  
Loss Given Default ◽  
Learning Techniques ◽  
Selection Algorithms

Credit risk assessment is of paramount importance in the financial industry. Machine learning techniques have been used successfully over the last two decades to predict the probability of loan default (PD). This way, credit decisions can be automated and risk can be reduced significantly. In the more recent parts, intensified regulatory requirements led to the need to include another parameter – loss given default (LGD), the share of the loan which cannot be recovered in case of loan default – in risk models. We aim to build a unified credit risk model by estimating both parameters jointly to estimate expected loss. A large, highdimensional, real world dataset is used to benchmark several combinations of classification, regression and feature selection algorithms. The results indicate that non-linear techniques work especially well to model expected loss.

scholarly journals Crypto Exchanges and Credit Risk: Modeling and Forecasting the Probability of Closure

2021 ◽  
Vol 14 (11) ◽  
pp. 516
Author(s):  
Dean Fantazzini ◽  
Raffaella Calabrese
Keyword(s):  
Machine Learning ◽  
Credit Risk ◽  
Credit Scoring ◽  
Machine Learning Techniques ◽  
Risk Modeling ◽  
Learning Techniques ◽  
Modeling And Forecasting ◽  
Out Of Sample ◽  
Significant Covariates ◽  
Unique Dataset

While there is increasing interest in crypto assets, the credit risk of these exchanges is still relatively unexplored. To fill this gap, we considered a unique dataset of 144 exchanges, active from the first quarter of 2018 to the first quarter of 2021. We analyzed the determinants surrounding the decision to close an exchange using credit scoring and machine learning techniques. Cybersecurity grades, having a public developer team, the age of the exchange, and the number of available traded cryptocurrencies are the main significant covariates across different model specifications. Both in-sample and out-of-sample analyzes confirm these findings. These results are robust in regard to the inclusion of additional variables, considering the country of registration of these exchanges and whether they are centralized or decentralized.

scholarly journals CREDIT RISK MODELLING FOR INDIAN DEBT SECURITIES USING MACHINE LEARNING

2021 ◽  
Vol 24 ◽  
pp. 107-128
Author(s):  
Charumathi Balakrishnan ◽  
Mangaiyarkarasi Thiagarajan
Keyword(s):  
Machine Learning ◽  
Credit Risk ◽  
Credit Rating ◽  
Risk Model ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Rating Agencies ◽  
Learning Techniques ◽  
Classical Models ◽  
Credit Risk Model

We develop a new credit risk model for Indian debt securities rated by major credit rating agencies in India using the ordinal logistic regression (OLR). The robustness of the model is tested by comparing it with classical models available for ratings prediction. We improved the model’s accuracy by using machine learning techniques, such as the artificial neural networks (ANN), support vector machines (SVM) and random forest (RF). We found that the accuracy of our model has improved from 68% using OLR to 82% when using ANN and above 90% when using SVM and RF.

scholarly journals Modeling of Bank Credit Risk Management Using the Cost Risk Model

2021 ◽  
Vol 14 (5) ◽  
pp. 211
Author(s):  
Iryna Yanenkova ◽  
Yuliia Nehoda ◽  
Svetlana Drobyazko ◽  
Andrii Zavhorodnii ◽  
Lyudmyla Berezovska
Keyword(s):  
Risk Management ◽  
Decision Support ◽  
Credit Risk ◽  
Risk Model ◽  
Bank Credit ◽  
Credit Risk Management ◽  
Loan Default ◽  
Increased Risk ◽  
Credit Portfolio ◽  
Cost Risk

This article deals with the issue of managing bank credit risk using a cost risk model. Modeling of bank credit risk management was proposed based on neural-cell technologies, which expand the possibilities of modeling complex objects and processes and provide high reliability of credit risk determination. The purpose of the article is to improve and develop methodical support and practical recommendations for reducing the level of risk based on the value-at-risk (VaR) methodology and its subsequent combination with methods of fuzzy programming and symbiotic methodical support. The model makes it possible to create decision support subsystems for nonperforming loan management based on the neuro-fuzzy approach. For this paper, economic and mathematical tools (based on the VaR methodology) were used, which made it possible to analyze and forecast the dynamics of overdue payment; assess the quality of the credit portfolio of the bank; determine possible trends in bank development. A scientific and practical approach is taken to assess and forecast the degree of credit problematicity by qualitative criteria using a mathematical model based on a fuzzy technology, which can forecast the increased risk of loan default at an early stage in the process of monitoring the loan portfolio and model forecasting changes in the degree of credit problematicity on change of indicators. A methodology is proposed for the analysis and forecasting of indicators of troubled loan debt, which should be implemented as software and included in the decision support system during the process of monitoring the risk of the bank’s credit portfolio.

scholarly journals Cocrystal Prediction Using Machine Learning Models and Descriptors

2021 ◽  
Vol 11 (3) ◽  
pp. 1323
Author(s):  
Medard Edmund Mswahili ◽  
Min-Jeong Lee ◽  
Gati Lother Martin ◽  
Junghyun Kim ◽  
Paul Kim ◽  
...  
Keyword(s):  
Machine Learning ◽  
Academic Research ◽  
Pharmaceutical Research ◽  
Machine Learning Techniques ◽  
Learning Models ◽  
Pharmaceutical Ingredients ◽  
Learning Techniques ◽  
Comparable Performance ◽  
Selection Algorithms ◽  
Machine Learning Models

Cocrystals are of much interest in industrial application as well as academic research, and screening of suitable coformers for active pharmaceutical ingredients is the most crucial and challenging step in cocrystal development. Recently, machine learning techniques are attracting researchers in many fields including pharmaceutical research such as quantitative structure-activity/property relationship. In this paper, we develop machine learning models to predict cocrystal formation. We extract descriptor values from simplified molecular-input line-entry system (SMILES) of compounds and compare the machine learning models by experiments with our collected data of 1476 instances. As a result, we found that artificial neural network shows great potential as it has the best accuracy, sensitivity, and F1 score. We also found that the model achieved comparable performance with about half of the descriptors chosen by feature selection algorithms. We believe that this will contribute to faster and more accurate cocrystal development.

Application of Ensemble Models in Credit Scoring Models

2018 ◽  
Vol 6 (2) ◽  
pp. 129-141
Author(s):  
Anjali Chopra ◽  
Priyanka Bhilare
Keyword(s):  
Decision Tree ◽  
Empirical Analysis ◽  
Credit Scoring ◽  
Bank Loan ◽  
Machine Learning Techniques ◽  
Risk Scores ◽  
Gradient Boosting ◽  
Loan Default ◽  
Linear Discriminant ◽  
Learning Techniques

Loan default is a serious problem in banking industries. Banking systems have strong processes in place for identification of customers with poor credit risk scores; however, most of the credit scoring models need to be constantly updated with newer variables and statistical techniques for improved accuracy. While totally eliminating default is almost impossible, loan risk teams, however, minimize the rate of default, thereby protecting banks from the adverse effects of loan default. Credit scoring models have used logistic regression and linear discriminant analysis for identification of potential defaulters. Newer and contemporary machine learning techniques have the ability to outperform classic old age techniques. This article aims to conduct empirical analysis on publically available bank loan dataset to study banking loan default using decision tree as the base learner and comparing it with ensemble tree learning techniques such as bagging, boosting, and random forests. The results of the empirical analysis suggest that the gradient boosting model outperforms the base decision tree learner, indicating that ensemble model works better than individual models. The study recommends that the risk team should adopt newer contemporary techniques to achieve better accuracy resulting in effective loan recovery strategies.

Leveraging Machine Learning in Financial Fraud Forensics in the Age of Cybersecurity

2022 ◽  
pp. 220-249
Author(s):  
Md Ariful Haque ◽  
Sachin Shetty
Keyword(s):  
Machine Learning ◽  
Financial Institutions ◽  
Machine Learning Techniques ◽  
Cyber Attack ◽  
Financial Industry ◽  
Security Breaches ◽  
Financial Gain ◽  
Learning Techniques ◽  
Machine Learning Model ◽  
Machine Learning Models

Financial sectors are lucrative cyber-attack targets because of their immediate financial gain. As a result, financial institutions face challenges in developing systems that can automatically identify security breaches and separate fraudulent transactions from legitimate transactions. Today, organizations widely use machine learning techniques to identify any fraudulent behavior in customers' transactions. However, machine learning techniques are often challenging because of financial institutions' confidentiality policy, leading to not sharing the customer transaction data. This chapter discusses some crucial challenges of handling cybersecurity and fraud in the financial industry and building machine learning-based models to address those challenges. The authors utilize an open-source e-commerce transaction dataset to illustrate the forensic processes by creating a machine learning model to classify fraudulent transactions. Overall, the chapter focuses on how the machine learning models can help detect and prevent fraudulent activities in the financial sector in the age of cybersecurity.

scholarly journals Credit Risk Assessment using Machine Learning Techniques

2019 ◽  
Vol 9 (1) ◽  
pp. 3482-3486
Keyword(s):  
Machine Learning ◽  
Risk Assessment ◽  
Random Forest ◽  
Credit Risk ◽  
Banking Sector ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Credit Risk Assessment ◽  
Learning Techniques ◽  
Cart Algorithm

Analysis of credit scoring is an effective credit risk assessment technique, which is one of the major research fields in the banking sector. Machine learning has a variety of applications in the banking sector and it has been widely used for data analysis. Modern techniques such as machine learning have provided a self-regulating process to analyze the data using classification techniques. The classification method is a supervised learning process in which the computer learns from the input data provided and makes use of this information to classify the new dataset. This research paper presents a comparison of various machine learning techniques used to evaluate the credit risk. A credit transaction that needs to be accepted or rejected is trained and implemented on the dataset using different machine learning algorithms. The techniques are implemented on the German credit dataset taken from UCI repository which has 1000 instances and 21 attributes, depending on which the transactions are either accepted or rejected. This paper compares algorithms such as Support Vector Network, Neural Network, Logistic Regression, Naive Bayes, Random Forest, and Classification and Regression Trees (CART) algorithm and the results obtained show that Random Forest algorithm was able to predict credit risk with higher accuracy

Sign in / Sign up

Export Citation Format

Share Document