Market Risk measurement in SimCorp Dimension
🪐 Market Risk Measurement models evolved significantly in SimCorp Dimension system. Dimension users have incredible set of choices: whether valuate Value-at-Risk (VaR) internally or import the risk figures from external providers (e.g. MSCI Inc. Risk Metrics).
💡 Many clients combine portfolio risk measurement and performance measurement reports together, what seems logical, taking into account that a majority of asset managers would like to know not only a nature of returns but also introduced risk level. Besides, for example Asset Manager application in Dimension has a possibility to integrate risk & performance figures as well.
🤔 How to choose between internal VaR calculation and import VaR from external data vendor? There are a few parameters to take into account:
1) a variety of instruments (e.g. complicated derivatives require more complex pricing to be available)
2) coverage rate of business operations, present in Dimension (i.e. if most of business operations are already there – you would benefit from having risk figures within the same system by re-using the available underlying data)
3) Market data costs (the expected costs of importing risk figures)
Thus, the 3 mentioned items from above should be enough to make such decision.
🧊 Regarding SimCorp Dimension risk measurement models themselves, I would suggest to keep in mind the following key points:
a) Dimension has a powerful risk measurement module, that supplies a wide variety of VaR-related ratios. Depending on the kind of investment strategy and type of the final user – you will be able to define which ratios matter to your specific situation
b) if you decided to go for in-house risk valuation models within SimCorp Dimension – that might require a solid expertise to get implemented, however it would be a profitable approach in the long term prospect anyway, taking into account the regular prices from external vendors for such VaR calculation services
c) for fixed income non-derivatives – you’ll mostly deal with interest rate risk and credit spread risks (what means you’d have to get ready to maintain many spread yield curves), whereas for equity-like non-derivatives you should remember that it is also possible to base VaR model on related index returns instead of individual equity returns
d) the VaR figures are usually not additive due to involved correlation matrixes, unless you start using a specific kind of VaR ratios (e.g. Incremental VaR ratios from Dimension system)
e) As long as you understand the theoretical valuation of instruments – you will have a clear picture of their market risk sources.