BoE published a working paper that examines the explainability of machine learning in finance through an application to default risk analysis. Machine-learning-based predictive techniques are seeing increased adoption in a number of domains, including finance. However, due to their complexity, their predictions are often difficult to explain and validate. This is sometimes referred to as machine learning’s "black box" problem. The paper studies a machine learning model to predict mortgage defaults and proposes a framework for addressing the "black box" problem present in some machine learning applications.
The paper addresses the explainability problem of artificial intelligence by studying the inputs and the outputs of a machine learning model.The approach is implemented by using the Quantitative Input Inﬂuence (QII) method of Datta et al (2016) in a real‑world example: a machine learning model to predict mortgage defaults. This method investigates the inputs and outputs of the model, but not its inner workings. It measures feature inﬂuences by intervening on inputs and estimating their Shapley values, representing the features’ average marginal contributions over all possible feature combinations. This method estimates key drivers of mortgage defaults such as the loan‑to‑value ratio and current interest rate, which are in line with the ﬁndings of the economics and ﬁnance literature. However, given the non‑linearity of machine learning model, explanations vary signiﬁcantly for different groups of loans. The study uses clustering methods to arrive at groups of explanations for different areas of the input space. Finally, the authors conduct simulations on data that the model has not been trained or tested on. The main contribution is to develop a systematic analytical framework that could be used for approaching explainability questions in real world ﬁnancial applications. The study concludes that notable model uncertainties do remain and stakeholders ought to be aware of these uncertainties.
The paper highlights regulators as one category of stakeholders that might be interested in the workings of machine learning model of mortgage defaults of a bank to assess the riskiness of its loan book. A regulator could usefully consider an influence-based explainability approach implemented by the bank. The paper also highlights that, in such situations, it is still difficult to estimate how a complex model would behave out of sample, for instance, in stress-test scenarios where inputs are deliberately stretched. The paper shows that explainable artificial intelligence tools are an important addition to the data science toolkit, as they allow for better quality assurance of black box machine learning models. These tools can usefully complement other aspects of quality assurance, including various ways of model performance testing, understanding the properties of the data set and domain knowledge.
Related Link: Working Paper
Keywords: Europe, UK, Banking, Credit Risk, Machine Learning, Artificial Intelligence, Regtech, Model Explainability, Mortgage Default, BoE
Previous ArticleNBB Launches the 2019 Stress Testing Exercise for Insurers
The European Commission (EC) published the Delegated Regulation 2021/1527 with regard to the regulatory technical standards for the contractual recognition of write down and conversion powers.
The Australian Prudential Regulation Authority (APRA) published a new set of frequently asked questions (FAQs) to provide guidance to authorized deposit-taking institutions on the interpretation of APS 120, the prudential standard on securitization.
The Single Resolution Board (SRB) published a Communication on the application of regulatory technical standard provisions on prior permission for reducing eligible liabilities instruments as of January 01, 2022.
The Australian Prudential Regulation Authority (APRA) published a new set of frequently asked questions (FAQs) to clarify the regulatory capital treatment of investments in the overseas deposit-taking and insurance subsidiaries.
The European Banking Authority (EBA) published the final report on the guidelines specifying the criteria to assess the exceptional cases when institutions exceed the large exposure limits and the time and measures needed for institutions to return to compliance.
The Prudential Regulation Authority (PRA) issued the policy statement PS20/21, which contains final rules for the application of existing consolidated prudential requirements to financial holding companies and mixed financial holding companies.
The European Banking Authority (EBA) revised the guidelines on stress tests to be conducted by the national deposit guarantee schemes under the Deposit Guarantee Schemes Directive (DGSD).
The European Commission (EC) announced that Nordea Bank has signed a guarantee agreement with the European Investment Bank (EIB) Group to support the sustainable transformation of businesses in the Nordics.
The Hong Kong Monetary Authority (HKMA) issued a circular, for all authorized institutions, to confirm its support of an information note that sets out various options available in the loan market for replacing USD LIBOR with the Secured Overnight Financing Rate (SOFR).
The Office of the Comptroller of the Currency (OCC) issued a new "Problem Bank Supervision" booklet of the Comptroller's Handbook. The booklet covers information on timely identification and rehabilitation of problem banks and their advanced supervision, enforcement, and resolution when conditions warrant.