Risk management framework
0 likes496 views
This document proposes a framework for real-time risk management. It outlines key concepts like portfolios, instruments, market data providers, risk analytics, and reports. It then describes key drivers like flexible portfolio specification and rapid integration of market data and models. Finally, it proposes an architecture with components for instrument services, portfolio services, market data validation, analytics, report generation, and management. The goal is to create a customizable, standards-based open source solution for financial risk management.
Risk management framework
- 2. This is my idea of a generic infrastructure for Realtime Risk (RR). Ideas are not entirely mine and have been gathered through several assignments I have held in Capital markets. If I had my way with sufficient funding this is how I would implement A financial risk management environment for an enterprise. I have tried to build a case around basic and essential requirements to achieve seamless integration within the orgamization and with external parties. customers have a different view of risk. I also believe it is hard to capture what exactly customers mean. Best practice in my opinion is to allow them to interact, discover and set their own risk parameters. Given all the data that is available from Yahoo and other internet outlets and given technologies like R and XML, Unix it is possible to engineer an easy to install, customize and create value. Free Software (GNU Style).
- 3. Key Concepts • Portfolios • Instruments • Market Data Providers • Risk Analytics • Reports • Interactive Experimentation
- 4. Key Drivers • Flexible PF Specification • Uniform Standard Instrument Definition • Rapid MDV Integration and Validation • Rapid Model/Analytic Integration • Report Generation • Delivery • Job Schedule and Monitoring • Security and Access Control
- 5. Instrument Services • Instrument Management – MultiAsset – Instrument Statistics+ • Standard Deviation, Closing Price – TimeSeries given Customer, DateRange – Daily,Monthly or other custom period • Beta against different benchmarks • Correlation and covariance with other instruments – Customer Independent – Customer Specific Attribution – Market Value and Customer Override – Notification and Access Control
- 6. Portfolio Services • MTM given date, Instrument Values • MTM Time Series, Return Time Series • Standard Deviation • Excess Return given Benchmark • Return per unit of Standard Deviation • Other statistics Sharpe,Sortino,Omega etc • MLE, VaR estimates, cVaR or other extensions • Customer Specific Uploads
- 7. MDV Service • Daily Upload • Symbology Mapping • Simple Validation • Error Report – No Change – AbnormalChange • Corporate Events • Analyst Override • Conflict Resolution
- 8. Analytics • Statistics given PF, Customer Specification • Key Statistics Persistence • AsOf Computation • Computation Given Alternate Valuation/Price • Exception Handling
- 9. Report Generation/Delivery • Flexible Specification • Flexible Rendering • Customization • Multiple Delivery Platforms • Security (Encryption) • Resource/Usage Meters Customer Specific Dashboards and OnDemand Services over secure channels
- 10. Management • Job Schedule Maintenance • Calendar, ReRun, Reporting • Access Control/Web Portals • Custom Deployment – Customer Can lease whole framework – Remote Administration – SAAS/AppServer
- 11. WAN Generic Architecture MDS RAS RPTS PFS SM RR ESB(XML/XBRL/JSON/FIX) RRR DBMS Customer Analyst SecSvc JMon RAS – Risk Analytic Server SM – Security master PFS – Position/Portfolio Server WAN RPTS – Report Server MDS – Market Data Server
- 12. Things To Do • Instrument Specification – Input/output, DBMS • PF Specification I/O, DBMS • Analytics Catalog – Reuse, I/O • Report Specification I/O, DBMS • Job Monitor Requirements • WAN Transport Specification (FTP/SMTP/HTTP) • Customer Specifications