際際滷

際際滷Share a Scribd company logo

LEI ON FIRE - April 2016

Download as PPTX, PDF
B
bfreeman1987

1) The document discusses the need for universal financial data standards like Legal Entity Identifiers (LEIs) to effectively monitor systemic risk. 2) The current LEI system faces challenges around data quality, lack of standardization between Local Operating Units, and not accommodating real-time aggregation needs. 3) A next-generation LEI system is proposed using blockchain technology run by Local Operating Units to provide global updates on a faster timescale while maintaining a complete historical record. This would help address current challenges around speed, scalability, and data quality.

1 of 11
Legal Entity Identifiers, Financial Reference Data and the need for Universality and Standards
The need for Quality Hierarchal LEI data, DTI?? and other emerging Universal Financial data standards (UTI, UPI, et al) are enabled thru advanced contextual graph models plus exploitation of Blockchain technological disruptive capability Ontology2 4/7/2016 Bill Freeman, Paul Houle , Allan Grody, patent pending
Daily, GlEIF concatenates and archives the data Financial Stability Board, CFTC,OFR, Treasury Reg Oversight Committee, Basil, EC, Global regulators Global Legal Entity Identifier Foundation Local Operating Units LOUs register entities and supply data files to GLEIF established oversees Global LEI System(GLEIS): A Work in ProcessCurrent State Under Stress Individual Records are supposed to be updated at least yearly, but lapse rates for such updating is 25% (although they should LEI records must be updated sooner for corporate actions to make the GLEIS relevant for use as golden copy for internal use by financial institutions Ultimately Needed : Real-time , streaming data analytics to monitor SIFI, trading , and the buildup of systemic risk capital adequecy LOUs publish daily updates, which are consolidated daily by the GLEIF. Process and data variables at LOUs problematic (i.e. ___________, ____________, __________) Note;add examples please A complete snapshot of the data is kept every day, giving a valuable record of system history Ontology2 4/7/2016
Current Challenges Uncertain use and value of LEIs in recording transactions in Trade Repositories (TRs) tens of billions of transactions have been sent to TRs with no way of validating LEIs usefulness and quality Different Internal Data and Communication Models at LOUs will prohibit advanced technology deployment if not standardized LEIs as mechanism to aggregate data on a hierarchical relationship basis is being put off to an uncertain outcome The use of the GLEIS for aggregating financial transaction data for systemic risk analysis, the key objective for regulators, is not yet accommodated in either the rule making or the design of the GLEIS The promise to financial industry members of straight-through-processing, significant infrastructure cost reduction and risk mitigation is not being realized, nor is it being identified in industry plans near term or in the future Little standard operating Procedure Manual Mutual Funds Ignored ( oretty much ) if not Challenging to understand the quality of information, sources and audit
Reference Data As Seen By Financial IT Trading software Entity data Product data Trade data Analytics Reporting Risk management Straight-through processing (risk elimination) Needs immediate reaction to changes affecting entities Analytics, Reporting (Risk management) Need to understand current state AND also what reference data was relevant to the trade at the time Ontology2 4/7/2016
Current LEI System is Not Fast,Scalable or Secure Update Rates Suffer 5 days of daily updates requires 5 copies of the data 5 days of hourly updates requires 120 copies of the data (Note: assume you are showing a single LOU example how many hours of operation to get to the calculations shown? Needs better example and hourly might not be fast enough to keep up with real-time trading and changing conditions under stress situations Ontology2 4/7/2016
Next-Generation LEI System Global Legal Entity Identifier Foundation manages Blockchain run collectively by Local Operating Units Blocks in chain represent changes to reference data published by Local Operating Units in conjunction with their registrants and their customers We get best of both worlds: Global update of reference data updated on a few second timescale AND Complete historical record at time of trade (keep block id for exact reference!) Ontology2 4/7/2016
Ontology2 Technology Enhances Blockchain Lambda Architecture operation (Superbatch ) to process updates in either a large batch (whole database) or in small batches (one block) Perfect for Blockchain! Ablility to accept or reject changes from blockchain to minimize the effect that incorrect or malicious data from blockchain impacts operations, reporting or analysis Semantic capability to represent, reason, and work with multiple representations of a situation (integrate blockchain-sourced reference data with legacy and future systems) Ontology2 4/7/2016
Index Cloud New Index Construction Never Conflicts With Production time Old index (multiple copies for throughput & availability) Source data Test Clone New Index Manage and recover resources Ontology2 4/7/2016
Intelligent Index ( ML ) plus Real-Time Index Build can be self correcting Effortless and efficient scalability Message Queue Bulk Data time stamped master data small real-time index large bulk index merger RESULTS Ontology2 4/7/2016
New approach to data management A FRAMEWORK FOR DATA QUALITY Multiple sources of instance data Facts classifications Reference data Examples Test Data Training Data Requirements Quality metrics Ontology2 4/7/2016

More Related Content

LEI ON FIRE - April 2016

  • 1. Legal Entity Identifiers, Financial Reference Data and the need for Universality and Standards
  • 2. The need for Quality Hierarchal LEI data, DTI?? and other emerging Universal Financial data standards (UTI, UPI, et al) are enabled thru advanced contextual graph models plus exploitation of Blockchain technological disruptive capability Ontology2 4/7/2016 Bill Freeman, Paul Houle , Allan Grody, patent pending
  • 3. Daily, GlEIF concatenates and archives the data Financial Stability Board, CFTC,OFR, Treasury Reg Oversight Committee, Basil, EC, Global regulators Global Legal Entity Identifier Foundation Local Operating Units LOUs register entities and supply data files to GLEIF established oversees Global LEI System(GLEIS): A Work in ProcessCurrent State Under Stress Individual Records are supposed to be updated at least yearly, but lapse rates for such updating is 25% (although they should LEI records must be updated sooner for corporate actions to make the GLEIS relevant for use as golden copy for internal use by financial institutions Ultimately Needed : Real-time , streaming data analytics to monitor SIFI, trading , and the buildup of systemic risk capital adequecy LOUs publish daily updates, which are consolidated daily by the GLEIF. Process and data variables at LOUs problematic (i.e. ___________, ____________, __________) Note;add examples please A complete snapshot of the data is kept every day, giving a valuable record of system history Ontology2 4/7/2016
  • 4. Current Challenges Uncertain use and value of LEIs in recording transactions in Trade Repositories (TRs) tens of billions of transactions have been sent to TRs with no way of validating LEIs usefulness and quality Different Internal Data and Communication Models at LOUs will prohibit advanced technology deployment if not standardized LEIs as mechanism to aggregate data on a hierarchical relationship basis is being put off to an uncertain outcome The use of the GLEIS for aggregating financial transaction data for systemic risk analysis, the key objective for regulators, is not yet accommodated in either the rule making or the design of the GLEIS The promise to financial industry members of straight-through-processing, significant infrastructure cost reduction and risk mitigation is not being realized, nor is it being identified in industry plans near term or in the future Little standard operating Procedure Manual Mutual Funds Ignored ( oretty much ) if not Challenging to understand the quality of information, sources and audit
  • 5. Reference Data As Seen By Financial IT Trading software Entity data Product data Trade data Analytics Reporting Risk management Straight-through processing (risk elimination) Needs immediate reaction to changes affecting entities Analytics, Reporting (Risk management) Need to understand current state AND also what reference data was relevant to the trade at the time Ontology2 4/7/2016
  • 6. Current LEI System is Not Fast,Scalable or Secure Update Rates Suffer 5 days of daily updates requires 5 copies of the data 5 days of hourly updates requires 120 copies of the data (Note: assume you are showing a single LOU example how many hours of operation to get to the calculations shown? Needs better example and hourly might not be fast enough to keep up with real-time trading and changing conditions under stress situations Ontology2 4/7/2016
  • 7. Next-Generation LEI System Global Legal Entity Identifier Foundation manages Blockchain run collectively by Local Operating Units Blocks in chain represent changes to reference data published by Local Operating Units in conjunction with their registrants and their customers We get best of both worlds: Global update of reference data updated on a few second timescale AND Complete historical record at time of trade (keep block id for exact reference!) Ontology2 4/7/2016
  • 8. Ontology2 Technology Enhances Blockchain Lambda Architecture operation (Superbatch ) to process updates in either a large batch (whole database) or in small batches (one block) Perfect for Blockchain! Ablility to accept or reject changes from blockchain to minimize the effect that incorrect or malicious data from blockchain impacts operations, reporting or analysis Semantic capability to represent, reason, and work with multiple representations of a situation (integrate blockchain-sourced reference data with legacy and future systems) Ontology2 4/7/2016
  • 9. Index Cloud New Index Construction Never Conflicts With Production time Old index (multiple copies for throughput & availability) Source data Test Clone New Index Manage and recover resources Ontology2 4/7/2016
  • 10. Intelligent Index ( ML ) plus Real-Time Index Build can be self correcting Effortless and efficient scalability Message Queue Bulk Data time stamped master data small real-time index large bulk index merger RESULTS Ontology2 4/7/2016
  • 11. New approach to data management A FRAMEWORK FOR DATA QUALITY Multiple sources of instance data Facts classifications Reference data Examples Test Data Training Data Requirements Quality metrics Ontology2 4/7/2016