Re: [Quantlib-users] ORE v10 released

[Forde S. <for...@gm...>]

I am assessing ORE for a client, and as part of this, I am compiling a list
of the major enhancements required to close regulatory gaps.

Could you review and comment on this initial text, which will be expanded
into a more comprehensive analysis?

ORE is an extremely powerful engine to either complement the internal model
team or provide the basis of a CCR platform for smaller institutions. It is
highly extensible and configurable.It has a tremendous foundation of data
handling, market and curve building, and analytic capabilities,
complemented by a comprehensive test suite. ORE focuses on market and
counterparty credit risk analytics and includes key aspects of the Basel
III framework, but it does not have all aspects, e.g., FRTB-specific
enhancements.

 There are some key functional and non-functional gaps and considerations
to address (not in order of priority), including:

   1.

   SA-CCR (Standardised Approach for Counterparty Credit Risk):While there
   are references in the user guide (page 4) to a future release of SA-CCR,
   the timing and commitment to this is unclear.
   2.

   ES (Expected Shortfall):With FRTB's introduction under Basel III,
   Expected Shortfall (ES) replaced VaR as the primary risk measure for the
   internal models approach. While VaR is still an important risk metric,
   implementing ES requires capturing the tail risk and different liquidity
   horizons for various asset classes and risk factors. There is a need to
   consider backtesting requirements, as ES introduces its challenges in that
   realm. ES means not just changing the formulae but also adopting a new risk
   mindset, focusing on tail risks and severe market shocks.
   3.

   SA-CVA (Standardised Approach for Credit Valuation Adjustment
   Risk):ORE's emphasis on xVA calculations provides a strong foundation, but
   the transition from a basic approach (e.g. KVA is calculated using the
   limited version fo BA-CVA) to a standardised approach like SA-CVA can be
   intricate. The details of risk factor modellability and NMRFs are
   fundamental to FRTB and will require a deep integration into ORE’s
   calculation mechanics. It's also worth noting that SA-CVA will require
   banks to possess a more granular level of data on their counterparties and
   credit risk mitigants.
   4.

   Enhanced data quality and traceability requirements. FRTB places a
   strong emphasis on data quality, and institutions are required to
   demonstrate the traceability of their risk data. This could necessitate not
   only infrastructure upgrades but also potential enhancements to ORE's data
   intake and validation processes. Additionally, there might be a need for
   more robust reporting functionalities to satisfy regulatory inquiries and
   audits.
   5.

   Scalability and Performance. Given the complexity of Basel III
   calculations, especially under approaches like SA-CCR or IMA of FRTB,
   there's a need for high computational efficiency. ORE's ability to scale
   and perform under these demands might be a consideration.
   6.

   Integration with other systems: If ORE is to be the foundation of a CCR
   platform, its ability to integrate with other bank systems (e.g., front
   office, collateral management, settlement, market data, broader risk
   systems, accounting systems) is crucial. The flow of data between these
   systems must be seamless. Capital markets are slowly standardising
   reporting (e.g. ISDA CRIF AND CDM) and there is a need to map the future
   direction against ORE’s capabilities.
   7.

   User Interface and Reporting: The system's ability to generate detailed
   reports, both for internal risk management and regulatory reporting, is
   vital. ORE's user interface might require enhancements to accommodate the
   added complexities of FRTB and SA-CCR.
   8.

   Future-proofing: Regulatory landscapes continue to evolve. Ensure any
   extensions built upon ORE are designed in a modular fashion, allowing for
   easier upgrades or adjustments as regulations change.

Thank you,

Forde





On Thu, 22 Jun 2023 at 20:29, Roland Lichters via QuantLib-users <
qua...@li...> wrote:

> Dear all,
>
>
>
> The 10th ORE release is out on https://github.com/OpenSourceRisk
>
>
>
> As announced, we have rolled out a range of hybrid products this time:
>
>    - Composite trades
>    - Collateralized Bond Obligations
>    - Generic Total Return Swaps and Contracts For Difference (CFDs)
>    accepting almost arbitrary underlying baskets
>    - Convertible Bonds and Asset Swapped Convertible Option Transactions
>    (ASCOTs)
>
>
>
> Moreover, we added to the analytics scope:
>
>    - ISDA's Standard Initial Margin Model (SIMM) with all versions from
>    inception to the most recent
>    - a proof-of-concept Credit Portfolio Model to construct portfolio
>    loss distributions due to credit migration, credit default and market moves
>    across cash products and derivatives
>
>
>
> See the release notes in
> https://github.com/OpenSourceRisk/Engine/blob/v1.8.10.0/News.txt
>
> or in the userguide.pdf at
> https://github.com/OpenSourceRisk/Engine/releases/tag/v1.8.10.0 for
> references to the related examples.
>
>
>
> And finally, the ORE Python module has been updated and can be installed
> with
>
>
>
> pip install open-source-risk-engine
>
>
>
> Please explore, all feedback is welcome!
>
>
>
> Best wishes,
>
> Roland
>
>
>
>
>
> *Roland Lichters*
>
> *Quantitative Services*
>
> [image: Logo Description automatically generated]
>
> Maurenbrecherstrasse 16, 47803 Krefeld, Deutschland
>
> office *+49* *2151 9284800 *mobile* +49 172 9985795*
>
> *rol...@ac... <rol...@ac...>*
>
> *acadia.inc <https://acadia.inc/>*
>
>
> [image: signature_2811593998]
> <https://www.youtube.com/channel/UCsMyFt94Jyfwo-ecLBpy5xw>[image:
> signature_1151196288] <https://twitter.com/AcadiaSoft_>[image:
> signature_3977098555] <https://www.linkedin.com/company/acadiasoft-inc>
>
>
>
>
>
>
>
> The information contained in this e-mail, and any attachment, is
> confidential and is intended solely for the use of the intended recipient.
> Access, copying or re-use of the e-mail or any attachment, or any
> information contained therein, by any other person is not authorized. If
> you are not the intended recipient please return the e-mail to the sender
> and delete it from your computer. The acadia.inc privacy policy is
> available on our website.
> _______________________________________________
> QuantLib-users mailing list
> Qua...@li...
> https://lists.sourceforge.net/lists/listinfo/quantlib-users
>