Search results

This chapter shows how to identify the characteristics of borrowers that are part of a credit scoring model. The credit risk scoring model is an important tool for evaluating credit risk associated with customer characteristics that affect defaults. This research was conducted at a financial institution, a subsidiary of a commercial bank in Indonesia, to answer the challenge of determining the feasibility of providing financing quickly and accurately. This model uses a logistic regression method based on customer data with indicators of demographic characteristics, assets, occupations, and financing payments. This study identifies nine variables that meet the goodness of fit criteria, which consist of WOE, IV, and p-value. The nine variables can be used as predictors of default probability: type of work, work experience, net finance value, tenor, car brand, asset price, percentage of down payment (DP), interest, and income. The results of the study form a risk assessment model to identify variables that have a significant effect on the probability of default.

We set out, in this paper, to extend the Das and Sundaram (2000) model as a means of simultaneously considering correlated default risk structure and counter-party risk. The multinomial model established by Kamrad and Ritchken (1991) is subsequently modified in order to facilitate the development of a computational algorithm for valuing two types of active credit derivatives, credit-spread options and default baskets. From our numerical examples, we find that along with the correlated default risk, the existence of counter-party risk results in a substantially lower valuation of credit derivatives. In addition, we find that different settings of the term structure of interest rate volatility also have a significant impact on the value of credit derivatives.

In fraudulent conveyance cases, plaintiffs allege that by entering into a complex leverage transaction, such as an LBO, a firm’s former owners ensured its subsequent collapse. Proving that the transaction rendered the firm insolvent may allow debtors (or their proxies) to claw back transfers made to former shareholders and others as part of the transaction.

Courts have recently questioned the robustness of the solvency evidence traditionally provided in such cases, claiming that traditional expert analyses (e.g., a discounted flow analysis) may suffer from hindsight (and other forms of) bias, and thus not reflect an accurate view of the firm’s insolvency prospects at the time of the challenged transfers. To address the issue, courts have recently suggested that experts should consider market evidence, such as the firm’s stock, bond, or credit default swap prices at the time of the challenged transaction. We review market-evidence-based approaches for determination of solvency in fraudulent conveyance cases.

We compare different methods of solvency determination that rely on market data. We discuss the pros and cons of these methods and illustrate the use of credit default swap spreads with a numerical example. Finally, we highlight the limitations of these methods.

If securities trade in efficient markets in which security prices quickly impound all available information, then such security prices provide an objective assessment of investors’ views of the firm’s future insolvency prospects at the time of challenged transfer, given contemporaneously available information. As we explain, using market data to analyze fraudulent conveyance claims or assess a firm’s solvency prospects is not as straightforward as some courts argue. To do so, an expert must first pick a particular credit risk model from a host of choices which links the market evidence (or security price) to the likelihood of future default. Then, to implement his chosen model, the expert must estimate various parameter input values at the time of the alleged fraudulent transfer. In this connection, it is important to note that each credit risk model rests on particular assumptions, and there are typically several ways in which a model’s key parameters may be empirically estimated. Such choices critically affect any conclusion about a firm’s future default prospects as of the date of an alleged fraudulent conveyance.

Simply using market evidence does not necessarily eliminate the question of bias in any analysis. The reliability of a plaintiff’s claims regarding fraudulent conveyance will depend on the reasonableness of the analysis used to tie the observed market evidence at the time of the alleged fraudulent transfer to default prospects of the firm.

There is a large body of literature in financial economics that examines the relationship between market data and the prospects of a firm’s future default. However, there is surprisingly little research tying that literature to the analysis of fraudulent conveyance claims. Our paper, in part, attempts to do so. We show that while market-based methods use the information contained in market prices, this information must be supplemented with assumptions and the conclusions of these methods critically depend on the assumption made.

The credit migration process contains important information about the dynamics of a firm's credit quality, therefore, it has a significant impact on its relevant credit derivatives. We present a jump diffusion approach to model the credit rating transitions which leads to a partial integro-differential equation (PIDE) formulation, with defaults and rating changes characterized by barrier crossings. Efficient and reliable numerical solutions are developed for the variable coefficient equation that result in good agreement with historical and market data, across all credit ratings. A simple adjustment in the credit index drift converts the model to be used in the risk-neutral setting, which makes it a valuable tool in credit derivative pricing.

Purpose: This chapter aims to examine machine learning (ML) models for predicting credit card fraud (CCF).

Need for the study: With the advance of technology, the world is increasingly relying on credit cards rather than cash in daily life. This creates a slew of new opportunities for fraudulent individuals to abuse these cards. As of December 2020, global card losses reached $28.65billion, up 2.9% from $27.85 billion in 2018, according to the Nilson 2019 research. To safeguard the safety of credit card users, the credit card issuer should include a service that protects customers from potential risks. CCF has become a severe threat as internet buying has grown. To this goal, various studies in the field of automatic and real-time fraud detection are required. Due to their advantageous properties, the most recent ones employ a variety of ML algorithms and techniques to construct a well-fitting model to detect fraudulent transactions. When it comes to recognising credit card risk is huge and high-dimensional data, feature selection (FS) is critical for improving classification accuracy and fraud detection.

Methodology/design/approach: The objectives of this chapter are to construct a new model for credit card fraud detection (CCFD) based on principal component analysis (PCA) for FS and using supervised ML techniques such as K-nearest neighbour (KNN), ridge classifier, gradient boosting, quadratic discriminant analysis, AdaBoost, and random forest for classification of fraudulent and legitimate transactions. When compared to earlier experiments, the suggested approach demonstrates a high capacity for detecting fraudulent transactions. To be more precise, our model’s resilience is constructed by integrating the power of PCA for determining the most useful predictive features. The experimental analysis was performed on German credit card and Taiwan credit card data sets.

Findings: The experimental findings revealed that the KNN achieved an accuracy of 96.29%, recall of 100%, and precision of 96.29%, which is the best performing model on the German data set. While the ridge classifier was the best performing model on Taiwan Credit data with an accuracy of 81.75%, recall of 34.89, and precision of 66.61%.

Practical implications: The poor performance of the models on the Taiwan data revealed that it is an imbalanced credit card data set. The comparison of our proposed models with state-of-the-art credit card ML models showed that our results were competitive.

This paper presents an empirical analysis of trade credit supplied by Japanese manufacturing firms and General Trading Companies. After reviewing major trade credit models and relevant Japanese literature, empirical tests examine the applicability of existing trade credit theories. Results indicate existing trade credit theory has little power to explain the level of trade credit supplied by Japanese firms. Instead, support is found for the information and risk sharing models in the Japanese keiretsu literature and the financial channeling process by which lending institutions supply General Trading Companies with liquidity that is, in turn, supplied to manufacturing firms.

Traditionally, loan officers use different credit scoring models to complement judgmental methods to classify consumer loan applications. This study explores the use of decision trees, AdaBoost, and support vector machines (SVMs) to identify potential bad loans. Our results show that AdaBoost does provide an improvement over simple decision trees as well as SVM models in predicting good credit clients and bad credit clients. To cross-validate our results, we use k-fold classification methodology.