xVA’s are one of the most challenging areas of finance in terms of computations but very much also in terms of data: massive sizes and numerous sources often require a heavy involvement from IT departments, potentially leading to longer delays and additional efforts from xVA desks and risk officers to process their data.
Limited data capabilities are a limiting factor for CCR and xVA management.
For instance, keeping nearly 2 trillions of Mark-to-Future intermediary points of all Monte Carlo paths:
Unfortunately those points are often discarded because that much data cannot be handled efficiently in existing set-up’s
ICA’s RAFAL results management database can absorb a massive Mark-to-Future cube at 350 Mo/s/node and run queries at 700Mo/s/node
Data can be handled by users with the flexibility of a pivot table with the capacity to access the most granular level (eg. deal leg)
RAFAL provides the ability to add proprietary analyses on the whole data set with customizable Python calculators allowing users to autonomously compute metrics as complex as CVA under CSA's
Counterparty Credit risk is better monitored and handled
The Credit Risk team can resolve their audit point on BSBC 239 (in accessing granular intermediary data, “what-if” scenarios) and can implement specific risk analysis
Risk officers enter a new dimension of opportunities to analyse credit risk with more speed and comprehensiveness, assess the relevance of limits, test macro hedges on tail scenarios, run back testing, clustering, what-if simulations...
Capital charge is much better challenged and anticipated (eg. through pre- and post-trade marginal impacts…)
would mean significant faster “what-if” simulations, improving data analysis and help challenge capital charges.
On top of being BCBS 239 compliant an easy access to those intermediary results would allow to solve several needs: cVaR attribution and explanation, backtesting...
+44 203 350 50 29 +33 1 53 65 35 97
SIGN UP FOR OUR LATEST UPDATES
ICA is a Capital Markets Fintech that empowers Big Data end-users to make advanced risk and business decisions.
For any GDPR information, please contact our Data Protection Officer: firstname.lastname@example.org