A fresh take on risk and valuation
We keep moving forward, opening
new doors and doing new things,
because we are curious, and curiosity keeps
leading us down new paths.
Machine Learning Model explainability –why is it important and methods to achieve it

Machine learning is today’s buzzword and it has gone through some phenomenal changes over the last few years. However, despite widespread adoption, machine learning models remain mostly black boxes. Hence it is essential to have techniques for explaining these black boxes in an interpretable manner. This article looks at the methods that are most used for interpreting machine learning models: SHAP, LIME and CP Profiles. It discusses advantages and drawbacks of each and shows how are they being used in practice.

Solvency 2020 review - EIOPA’s opinion on recovery and resolution

The 2008 financial crisis has shown that it is vital to have a good recovery and resolution policy. Because the European-level policy on resolution and recovery of the insurers is insufficient, some national authorities have adopted their own recovery and resolution policies. However this has led to a fragmented regulatory landscape across the EU. Therefore, the European Commission was requested to create a harmonised framework. This article explains eight proposed recommendations on the future amendments to the Solvency II regulation published on that topic by EIOPA In the Consultation Paper on Solvency 2 review 2020.

New definition of default: tips to simulate the future impact!

Every bank is highly interested in predicting the impact that the application of the new default definition would have on various aspects. That prediction can be achieved via a comprehensive simulation of the default under the new definition on the available historical data of the credit portfolio. Knowing the broad impact of the new definition on capital requirements, IFRS 9 models, precision is highly recommended for the sake of planning of own funds requirements.

2020 Solvency II review – Technical Provisions and SCR

The 2020 Solvency II review intends to bring about several changes to the Solvency II Framework Directive. It follows the 2018 Solvency II interim review, which amended the Solvency II Delegated Acts. In this article, we focus on the EIOPA’s opinions on technical provisions. For now, we are excluding Long Term Guarantee (LTG) measures, a topic which deserves an article all to itself.


Following the introduction of Solvency 2, the European Commission and Council have decided to continue their efforts to develop a common regulatory framework for the financial industry to strengthen its governance. On 13th January 2017, the regulators laid down a new milestone in the pensions market with IORP2, the new directive that replaced the existing and much amended directive 2003/41/CE. This new directive had to be transposed into national laws before 13th January 2019.

ISDA SIMM for non-cleared options

Equity options are excluded the collection of IM until January 2021 in the EU. “The option seller should be able to choose not to collect initial or variation margins for these types of OTC derivatives as long as the option seller is not exposed to any credit risk. The counterparty paying the premium (‘option buyer’) should however collect both initial and variation margins”. Thereby, even if the option seller does not face any potential future exposure with the premium paid, the option buyer is still required to post the IM and VM, unless he is able to isolate this trade from other potential future exposures with the option seller.

Like love, risk appetite is all around

Truth be told, from a regulatory perspective, there is much more of the latter going on. Risk appetite is now widely recognised to be the main maker – or breaker – of success for financial institutions. As post mortem examinations of failed companies showed, a poor definition of risk appetite potentially leads to multiple errors in the course of business, such as taking unwarranted risks, resorting to short-term funding while taking long-term commitments, or to a general lack of alignment towards risk taking within the firm. These risk management mistakes could already prove to be fatal to the organization in isolation; if they happen simultaneously within an institution, demise is most certainly unavoidable.

Private Equity: adjusting valuation output

Private equity valuation considering illiquid stakes, companies at an early stage of development for which future cash-flows are very uncertain, or companies past the startup phase, but without significant revenues and with negative EBITDA. In this article, these particularities are highlighted, as well as the possible adjustments and methods that can be used when valuing private companies. We assume that, since you are at the stage where you are looking for elements to adjust your valuation output, you already are at an advanced stage of your valuation process.

Basel IV - data from a bank's perspective

This blog discusses the impact that the December 2017 Basel reforms will have on the way banking institutions are going to use their data when constructing their credit risk models. Whilst these changes are not going to impact every financial institution in exactly the same way and the bigger institutions that make use of IRB approach will be affected disproportionately harder, this article provides a description of all major Basel IV data elements that all banking institutions will have to account for and in what way. These elements include: External Ratings, Collaterals Sourcing, Credit Conversion Factor (CCF), SME Indicator, Revolver/Transactor Indicator and others. Coordinates to the specific regulatory pieces are provided if you are interested in exploring any particular topic in greater detail.

The Language War in Credit Risk Modelling: SAS, Python or R?

The three languages were compared using a simple setup, as close as possible to a real-life situation. The exercise consisted in calibrating a logistic regression to identify loans likely not to be repaid on time in a sample dataset. The choice of logistic regression was driven by the fact that it is a simple but powerful approach still widely used in the industry.