Understanding Model Risk Management - Model Rationalization in Financial Industry.

Hevel Jean-Baptiste,Lixin Tao,Meikang Qiu,Keke Gai
DOI: https://doi.org/10.1109/cscloud.2015.68
2015-01-01
Abstract:Building a good financial Model Risk Management(MRM) system requires advanced skills in technology, econometrics and financial businesses. Very often we have businesses that have different models with similar functionality. The question that we are trying to answer in the paper is how to rationalize different models within an Enterprise Risk Management(ERM) of in organization. It is a dilemma that Risk Management Practitioners are struggling with. Very often they are looking for someone who can understand their business ideas and transform them to technology solutions or come up with technology solutions to support their MRM ideas. While quantitative Analysts are very critical in this type of project, very often they fail in the software development skills that are required. We are witnesses these struggles where some model risk leaders fail to find some good technologists or econometricians to implement their risk management philosophy. Also we have cases where technologists or econometricians cannot receive proper requirements to implement the functionality that the business needs. Here we are trying to understand how we can rationalize on different models that have some common functionality. For example how to look at or consolidate the functionality of the following models, such as Probability of Default (PD), Loss Given Default (LGD), Allowance for Loan and Lease Losses (ALLL), Stress Testing (ST) and others Financial Models. In this paper, we will present a framework for Model Rationalization or how to consolidate models. Also we will present a methodology that can be used to test a model that has been consolidated through our framework, Model Rationalization.
What problem does this paper attempt to address?