Default definition selection for credit scoring

Harris, Terry

doi:10.5430/air.v2n4p49

Cited by 7 publications

(4 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using different default definitions were first explored in Harris (2013b) and Harris (2013a), wherein the model accuracy of support vector machines predicting default are studied whilst employing various default definitions. However, while optimising accuracy is certainly worthwhile, the implications of variable default definitions for overall profitability are less clear.…”

Section: A Defining Background On Loan Defaultmentioning

confidence: 99%

A procedure for loss-optimising default definitions across simulated credit risk scenarios

Botha¹,

Beyers²,

Villiers³

2019

Preprint

View full text Add to dashboard Cite

A new procedure is presented for the objective comparison and evaluation of default definitions. This allows the lender to find a default threshold at which the financial loss of a loan portfolio is minimised, in accordance with Basel II. Alternative delinquency measures, other than simply measuring payments in arrears, can also be evaluated using this optimisation procedure. Furthermore, a simulation study is performed in testing the procedure from 'first principles' across a wide range of credit risk scenarios. Specifically, three probabilistic techniques are used to generate cash flows, while the parameters of each are varied, as part of the simulation study. The results show that loss minima can exist for a select range of credit risk profiles, which suggests that the loss optimisation of default thresholds can become a viable practice. The default decision is therefore framed anew as an optimisation problem in choosing a default threshold that is neither too early nor too late in loan life. These results also challenges current practices wherein default is pragmatically defined as '90 days past due', with little objective evidence for its overall suitability or financial impact, at least beyond flawed roll rate analyses or a regulator's decree.

show abstract

Section: A Defining Background On Loan Defaultmentioning

confidence: 99%

A procedure for loss-optimising default definitions across simulated credit risk scenarios

Botha¹,

Beyers²,

Villiers³

2019

Preprint

View full text Add to dashboard Cite

show abstract

“…The present study is closest in design to the work of Harris (2013a), Harris (2013b), Botha et al (2021), and Botha et al (2022). In particular, Harris (2013a) proposed an algorithm (using random forests with data from Barbados) that yields the 'best' default definition based on maximising prediction accuracy. When measured in days past due (DPD), these definitions included: 30, 60, and 90 days.…”

Section: Introductionmentioning

confidence: 99%

Defining and comparing SICR-events for classifying impaired loans under IFRS 9

Botha¹,

Oberholzer²,

Larney³

et al. 2023

Preprint

View full text Add to dashboard Cite

The IFRS 9 accounting standard requires the prediction of credit deterioration in financial instruments, i.e., significant increases in credit risk (SICR). However, the definition of such a SICR-event is inherently ambiguous, given its reliance on comparing two subsequent estimates of default risk against some arbitrary threshold. We examine the shortcomings of this approach and propose an alternative framework for generating SICR-definitions, based on three parameters: delinquency, stickiness, and the outcome period. Having varied these parameters, we obtain 27 unique SICR-definitions and fit logistic regression models accordingly using rich South African mortgage data; itself containing various macroeconomic and obligor-specific input variables. This new SICR-modelling approach is demonstrated by analysing the resulting portfolio-level SICR-rates (of each SICR-definition) on their stability over time and their responsiveness to economic downturns. At the account-level, we compare both the accuracy and flexibility of the SICR-predictions across all SICR-definitions, and discover several interesting trends during this process. These trends form a rudimentary expert system for selecting the three parameters optimally, as demonstrated in our recommendations for defining SICR-events. In summary, our work can guide the formulation, testing, and modelling of any SICR-definition, thereby promoting the timeous recognition of credit losses; the main imperative of IFRS 9.

show abstract

“…On the other hand, Random forest is a proven method for analyzing the influence of explanatory variables upon explained variable. [11,12] As described in Section 3, the authors propose to configure initial state of Bayesian network leveraging the result of Random forest analysis. The initial state consists of a few nodes around the target node and several edges between these nodes and the target.…”

Section: Introductionmentioning

confidence: 99%

Properly initialized Bayesian Network for decision making leveraging random forest

Iwakami

Takuma²,

Iwashita³

2020

AIR

View full text Add to dashboard Cite

Bayesian network is one of major methods for probabilistic inference among items. But if it contains particular targeting node and other explanatory nodes for decision making, for example how to select suitable appealing keywords to make customers like a product, edges around the target should be counted with more importance than those among others while constructing the network. In order to achieve this adjustment, this study proposes to configure initial state consisting of a few nodes and their edges connected with the target. The initial state is obtained by leveraging Random forest which is a proven method for decision making. Initial nodes are extracted by measuring mean decrease of Gini coefficient calculated with decision trees of Random forest. Initial edges are designated by comparing Kullback-Leibler divergences of conditional probability distribution among nodes which are corresponding to edge directions. Through an actual experiment, this method is confirmed to be effective for adjusting Bayesian network in decision making. This approach is especially useful for business scenes, such as selecting preferable keywords for appealing products over SNS.

show abstract

Default definition selection for credit scoring

Cited by 7 publications

References 23 publications

A procedure for loss-optimising default definitions across simulated credit risk scenarios

A procedure for loss-optimising default definitions across simulated credit risk scenarios

Defining and comparing SICR-events for classifying impaired loans under IFRS 9

Properly initialized Bayesian Network for decision making leveraging random forest

Contact Info

Product

Resources

About