Model, model risk and running effective model management programs

thumbnail of model-risk-and-running-effective Original by Cognizant, 2015, 9 pages 

This summary note was Posted on

There is a case to consider MRM as integrated part of business strategy rather than from a regulatory compliance perspective alone

  • The SR11-07 guideline (2011) refers to a model as a quantitative method, system or approach that applies statistical, economic, financial or mathematical theories , techniques and assumptions to process input data into quantitative estimates
  • Broader scope encompasses model development, implementation and use as well as governance and controls related to models (not only model validation)
  • Care should be taken that policies are clear and unambiguous
  • An action plan should be laid for addressing any of the gaps
  • Under documentation of models prevents the institutionalization of knowledge, which is the biggest risk banks face since the knowledge of the model moves with the modeler
  • An inventory should capture complete metadata about a model (owner, risk of the model, dates of the model build and reviews, implementation date, information on all the assumptions, linked models and products)

Sources of model risk (unsystematic)

Model errors
  • Mathematical errors, approximations or assumptions that are misleading or inappropriate
Data errors
  • Missing or incomplete, duplicated data
  • Inaccurate data
  • Outdated data
  • Asynchronous data
  • Misinterpreted data
  • Need to be able to trace data lineage (where does it come from, who owns it, costs of making changes)
  • Data model dependencies
Implementation errors
  • Model translation into framework
  • Errors in the code
Usage errors
  • Usage of the model instead of another one because it returns better results

Financial modeler’s manifesto

  • by E. Derman and P. Willmott, 2009
  • I will remember that I didn’t make the world, and it doesn’t satisfy my equations
  • Though I will use models boldly toe simulate value, I will not be overly impressed by mathematics
  • I will never sacrifice reality for elegance without explaining why I have done so
  • Nor will I give the people who use my model false comfort about its accuracy. Instead I will make explicit its assumptions and oversights
  • I understand that my work may have enormous effects on society and the economy, many of them beyond my comprehension