BCBS 239 Risk Data Aggregation: The data scope is enormous - where & how to begin? The Basel Committee on Banking Supervision (BCBS) has issued Regulation 239; a mandatory requirement for banks to aggregate risk data and provide comprehensive reporting. A significant upgrade of Risk IT and Systems is imperative
Model DR dynamically builds an alternative data viewpoint
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
BCBS 239 Risk Data Aggregation:
The data scope is enormous
- where & how to begin?
The Basel Committee on Banking Supervision (BCBS)
has issued Regulation 239; a mandatory requirement for
banks to aggregate risk data and provide
comprehensive reporting.
A significant upgrade of Risk IT and Systems is
imperative
Panoptic & Congruent
Senior management requires a panoptic view of the banks
dataset
To be panoptic the data set needs to demonstrate
congruence.
• Congruent data structures
allow viewing flexibility – any
data request can be satisfied
• Hard coded data solutions do
not allow a panoptic data view
– we want to ask for anything
BCBS 239 compliance requires an
Action Plan
2013-09 How to industrialize business reporting3
G-sib banks must be BCBS 239 compliant by Jan 2016
• The Model DR Action Plan introduces three new concepts
– Industrialisation
– Global language
– Congruent panoply
Model DR does not
provide a bolt on
solution. This is
baking the solution
into your business
Assembling the data is too big a job for one
Project Team
• What is the data set- Assets & Liabilities - multi currency
- Counterparty Credit
- Value at Risk (VAR)
- Derivatives…….
• Data cascades, horizontally and vertically, in multifarious but interconnected hierarchies, across the bank
• One data aggregation model is required, using a common language, to sit over the various internal system architectures
• Model DR provides a congruent, panoptic data view from which data may be aggregated and reports assembled
• Model DR breaks data into atomic elements, providing the tools for flexible restructuring
2013-09 Datapoint modelling and its opportunity 4
Many pieces of data, but a universal
way to view it
Regulatory Demand for Data –
when & what
• Banks cannot predict from what view point regulators may demand reports
and data
• Model DR provides the tools to build a new viewpoint ‘on the fly’
Regulatory scenario - Greece defaults on its bank debt, as Europe slides into
recession. What is exposure to German/ French banks – debt, derivatives,
Greek counterparties?
2013-09 How to industrialize business reporting5
Aggregation Model
Model DR - congruent panoptic view
Foreign
Currency
Loans
Derivatives SecuritiesCounterparty
credit
Model DR Concept: Industrialization
• Model DR enables banks to open up and examine existing systems by
mapping to the DPM
• The industrialization process automates, provides new tools and makes
systems more accessible
• Each project team is given tools & training
• Progress is measured and project teams are accountable• Are teams following the process correctly
• Are they mapping to the global language
• Are they exposing data in the agreed way
• The industrialization process is too large for one project
management team – instead responsibilities are shared by project
groups
• Data point modelling provides the methodology to reverse and
forward engineer Model DR production design into the global
language
2013-09 How to industrialize business reporting6
Model DR Concept: The global
language
• FIBO (Finance Industry Business Ontology)
- an industry initiative to define finance industry terms
- provides for data congruence
- facilitates data integration, supports business process automation and enables consolidated views across the financial industry
- driven by Dodd Frank regulatory requirements for improvements in data quality and transparency
- FIBO ontology defines financial terms and concepts without ambiguity
• XBRL (eXtensible Business Reporting Language)
- a business language used by major regulators to standardise financial reporting terms
- XBRL allows universal communication through metadata taxonomies which capture and define financial concepts,
terms & relationships
2013-09 How to industrialize business reporting7
Model DR Concept : data set must be
congruent to model
2013-09 How to industrialize business reporting8
• Model DR provides all users with a single, enterprise view of portfolio
risk and exposure. It is not a data warehouse
• Data Point Model (DPM) provides a literal representation of the data,
identifying all the business concepts and relations, as well as validation
rules. It contains all the relevant technical specifications necessary for
developing an IT reporting solution.
The Data Point Model
2013-09 How to industrialize business reporting9
Value
Value Set
Data point
Aspect
CubeBusiness class
Resource link
ResourceAxis
Axis coord.
Table
The data point meta model
has a dozen main concepts
Cube region
ValueContext
Many 1
Reportable Aspect
X Axis Aspect
X Axis Value Set
X Axis Coordinate Aspect Values
Report name
Y Axis Aspect
Y Axis Value Set Name
Y Axis Coordinate Aspect Values
Reportable Aspect Values
Package name
Example 1: Reverse engineer a report
into a Data Point Model
Business entity
Attribute Adaptors
Adaptors
Class level adaptor
Resources
Class level Aspect
Attribute level Aspects
Attribute level Value Sets
Example 2: Reverse engineer a data
base into a DPM
Example 3: Designing a new viewpoint = wiring up
DPM to DPM (with comparability)
2013-09 How to industrialize business reporting12
A third DPM
wired together
from 2 existing
DPM
The Model DR product
offering: • Dynamically builds an alternate data viewpoint
• Allows forward and reverse engineering of data in existing system architecture
• Allows universal structuring and data reassembly at an atomic level
• Provides the tools for BCBS strategy development
• TTE -Tooling for industrialization
• TTE -Training : • BCBS specification
• Developing the global language (Meta modelling)
• Designing and building viewpoints (Data Point Modelling)