XBRL Taxonomy Extension Comparability Issues And Potential Semantic Web Solutions Ashu Bhatnagar Good Morning Research – a...
<ul><li>Taxonomy Extensions Architecture and Data Organization  </li></ul><ul><li>Comparability Issues due to Extensions f...
XBRL Architecture Source: IFRS Taxonomy Guide 2009
Wall Street View - Data Will Be A Commodity Strategic View SEC Filings (10-K/10Q) Earnings / Call Reports Annual Reports F...
XBRL Taxonomy Extension  Comparability Issues
Taxonomy Extensions – What and Why? Taxonomy Extensions Core Taxonomy <ul><li>Types: </li></ul><ul><li>Entity-Specific Ext...
Source :Cliff Binstock, XBRL Cloud,  http://edgardashboard.xbrlcloud.com/edgar-rss-index.html Identifying Comparability Is...
Raw Data (Excel, Text) XBRL Files 1 Data (files) Aggregation Raw Data (Excel, Text) XBRL Files 2 Raw Data (Excel, Text) XB...
Comparability Issues for Filers <ul><li>Filers are driven by regulatory compliance and meeting deadlines with error free f...
Comparability Issues for Regulators - 1 <ul><li>Taxonomy extensions -  </li></ul><ul><li>Help with more precise </li></ul>...
Comparability Issues for Regulators - 2 <ul><li>Taxonomy extensions reduce standardization and make interoperability more ...
Comparability Issues for Technology Vendors Before (HTML, PDF) After (XBRL ZIP) <ul><li>XBRL Taxonomy extension is a manua...
Comparability Issues for Data Aggregators <ul><li>Data aggregators and distributors have not embraced XBRL in  </li></ul><...
Comparability Issues for Sell Side Inst. Investors Wall Street Sell-Side Experience: <ul><li>XBRL is still early in its ad...
Comparability Issues for Buy Side Investors Australian Hedge Fund Experience: <ul><li>Trading on multiple exchanges – Aust...
Comparability Issues for Retail Investors <ul><li>No direct impact on Retail investors. </li></ul><ul><li>Yet. </li></ul><...
Comparability Issue Due to Trust Consideration
Taxonomy Extensions Comparability Issues  Authored and Edited  by Selected Experts Authored and Edited by experts  &  non-...
Potential Semantic Web Solutions
Semantic Web & XBRL Architectures Semantic Web (Layered Cake Diagram)
W IKI g o o g l e PEDI A XB RL Process Technology Process Human Machine Human Man-Machine-Man Expert System Text Data In C...
POC: RDF/XBRL/Sparql Mashup As Data Quality Heat Map <ul><li>Proof of Concept: </li></ul><ul><li>Data Quality Heat Map </l...
Questions? [email_address] Thank You
Upcoming SlideShare
Loading in …5
×

W3C-XBRL International Workshop 2009

654 views

Published on

W3C-XBRL International Workshop 2009, hosted by FDIC, Arlington, VA, Oct. 5-6, 2009

0 Comments
2 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
654
On SlideShare
0
From Embeds
0
Number of Embeds
24
Actions
Shares
0
Downloads
0
Comments
0
Likes
2
Embeds 0
No embeds

No notes for slide

W3C-XBRL International Workshop 2009

  1. 1. XBRL Taxonomy Extension Comparability Issues And Potential Semantic Web Solutions Ashu Bhatnagar Good Morning Research – a Softpark Service Workshop on Improving Access - Financial Data on the Web Co-organized by W3C and XBRL International, Inc, and hosted by FDIC, Arlington, Virginia USA 5 - 6 October 2009
  2. 2. <ul><li>Taxonomy Extensions Architecture and Data Organization </li></ul><ul><li>Comparability Issues due to Extensions for </li></ul><ul><ul><li>Filers </li></ul></ul><ul><ul><li>Regulators </li></ul></ul><ul><ul><li>Technology Vendors </li></ul></ul><ul><ul><li>Data Aggregators and Distributors </li></ul></ul><ul><ul><li>Institutional Investors (Sell-Side) </li></ul></ul><ul><ul><ul><ul><li>Wall Street experience </li></ul></ul></ul></ul><ul><ul><li>Institutional Investors (Buy-Side) </li></ul></ul><ul><ul><ul><ul><li>Australian Hedge Fund experience </li></ul></ul></ul></ul><ul><ul><li>Retail Investors </li></ul></ul><ul><ul><ul><ul><li>Bombay Stock Exchange experience </li></ul></ul></ul></ul><ul><li>Comparability Issue due to Trust Considerations </li></ul><ul><li>Potential Semantic Web Solutions </li></ul>Topics
  3. 3. XBRL Architecture Source: IFRS Taxonomy Guide 2009
  4. 4. Wall Street View - Data Will Be A Commodity Strategic View SEC Filings (10-K/10Q) Earnings / Call Reports Annual Reports Filers/Companies Data Providers/ Sell-side Analysts Valuation/Forecast Models Stock Recommendations Buy-side Analysts Consensus Estimates Trading Strategies Data Aggregators Ratings Screeners Comps Back - Test Portfolio Construction Risk Management Investor Presentations Sanjiv Das Former Managing Director Morgan Stanley Expert View “ Data will be a commodity , but organization of data may NOT be a commodity” Data Information Analysis Insights Long Tail of Data Organization World of Accounting (GAAP) <ul><li>50 to 100 common Financial Metrics for </li></ul><ul><li>Income Statement </li></ul><ul><li>Balance Sheet </li></ul><ul><li>Cash-flow </li></ul>World of Analysts (Industry Specific Taxonomies) World of Investors (Folksonomy) <ul><li>Additional Unlimited , unstructured ‘localized’ synonyms for many Financial Metrics for </li></ul><ul><li>compliance with country-specific regulators </li></ul><ul><li>proprietary financial modeling frameworks </li></ul><ul><li>lack of any appropriate metrics in GAAP or </li></ul><ul><li>readily available taxonomies, resulting in – </li></ul><ul><li>Line 6 Items on 10-K, MDAs, and Footnotes </li></ul>GoodMorning Research LongTail View Wiki-Tagged Data Organization Process <ul><li>Additional 2,000 + unique Financial Metrics for </li></ul><ul><li>Financial Modeling </li></ul><ul><li>Valuation </li></ul><ul><li>Forecasts, </li></ul><ul><li>Trading Strategies </li></ul>
  5. 5. XBRL Taxonomy Extension Comparability Issues
  6. 6. Taxonomy Extensions – What and Why? Taxonomy Extensions Core Taxonomy <ul><li>Types: </li></ul><ul><li>Entity-Specific Extensions </li></ul><ul><ul><li>Adding extended link roles (ELRs) </li></ul></ul><ul><ul><li>Adding concepts </li></ul></ul><ul><ul><li>Adding links hierarchies </li></ul></ul><ul><ul><li>Adding links dimensions </li></ul></ul><ul><li>Receiving FIs Extensions </li></ul><ul><ul><li>Reuse of concepts in other reports </li></ul></ul><ul><ul><li>Creating new hierarchies </li></ul></ul><ul><ul><li>Adding new disclosures, etc. </li></ul></ul>Main objective for extensions is to increase the usability of Core Taxonomy Reference: IFRS Taxonomy Guide 2009
  7. 7. Source :Cliff Binstock, XBRL Cloud, http://edgardashboard.xbrlcloud.com/edgar-rss-index.html Identifying Comparability Issues
  8. 8. Raw Data (Excel, Text) XBRL Files 1 Data (files) Aggregation Raw Data (Excel, Text) XBRL Files 2 Raw Data (Excel, Text) XBRL Files n Data Extraction/ Normalization SQL Database Filers Data Aggregation & Distribution Comps Screens Valuations Portfolio Construction Risk Models Analytics Non-XBRL Data Market Data, Transactions Data Fixed Income Data Analysts Forecast & Historical Data Currency Data Information Supply Chain
  9. 9. Comparability Issues for Filers <ul><li>Filers are driven by regulatory compliance and meeting deadlines with error free filings </li></ul><ul><li>Comparability issue are primarily with their own previous filings </li></ul><ul><li>Taxonomy extension is onerous work, but preferred for legally precise disclosures </li></ul>
  10. 10. Comparability Issues for Regulators - 1 <ul><li>Taxonomy extensions - </li></ul><ul><li>Help with more precise </li></ul><ul><li>and detailed/transparent data, for better analytics </li></ul><ul><li>(such as banking regulations, capital to risk weighted asset ratio, Tier I capital ratio, compliance with Basel II, etc.) </li></ul><ul><li>- Makes it more difficult to compare apples-to-apples </li></ul>
  11. 11. Comparability Issues for Regulators - 2 <ul><li>Taxonomy extensions reduce standardization and make interoperability more difficult </li></ul><ul><li>Comparing same company data from quarter to quarter and analyzing re-statements is more difficult. </li></ul><ul><li>Adopting a maintenance program to update the existing taxonomy on demand and managed by the XBRL jurisdiction amounts to increased work for the taxonomy management (definition of acceptance criteria, editing, testing etc.) </li></ul><ul><li>Countries like India prohibit XBRL Extensions and use a pre-tagged XBRL blank form to be filled by the filers </li></ul><ul><li>Italy uses a similar approach with an option to submit an additional PDF file for more detailed financials with taxonomy extensions </li></ul>
  12. 12. Comparability Issues for Technology Vendors Before (HTML, PDF) After (XBRL ZIP) <ul><li>XBRL Taxonomy extension is a manual process requiring accounting skills </li></ul><ul><li>Few technologists on XBRL projects have accounting domain knowledge </li></ul><ul><li>Few Accounting experts have technology domain knowledge </li></ul><ul><li>Communication of accounting terms, adjustments, and technology options </li></ul><ul><li>between two expert areas is generally challenging </li></ul>
  13. 13. Comparability Issues for Data Aggregators <ul><li>Data aggregators and distributors have not embraced XBRL in </li></ul><ul><li>any significant way </li></ul><ul><li>When they do, taxonomy extensions will help them in extracting deep and transparent insights for Alpha driven analytic tools </li></ul><ul><li>Taxonomy extensions will amount to more complex and expensive </li></ul><ul><li>technology development </li></ul>
  14. 14. Comparability Issues for Sell Side Inst. Investors Wall Street Sell-Side Experience: <ul><li>XBRL is still early in its adoption on Wall Street </li></ul><ul><li>Sell-side research analysts would prefer XBRL taxonomy extensions </li></ul><ul><li>for deeper insights and increased transparency reasons </li></ul><ul><li>Comparability Issues due to proprietary Accounting and Valuation </li></ul><ul><li>frameworks and taxonomies (not XBRL) </li></ul><ul><ul><li>Proprietary accounting adjustments for Pensions, Leases, Valuations </li></ul></ul><ul><ul><li>Business rules for differences in fiscal period end dates </li></ul></ul><ul><ul><li>Adjustments for hyper-inflationary economies </li></ul></ul><ul><li>Comps analysis requires co-mingling reported data with historical data, </li></ul><ul><li>forecast data, index data, currency rates and live price data </li></ul>
  15. 15. Comparability Issues for Buy Side Investors Australian Hedge Fund Experience: <ul><li>Trading on multiple exchanges – Australia, Hong Kong and Singapore </li></ul><ul><li>Hardly any tools available for consuming XBRL data for Portfolio </li></ul><ul><li>Construction and Risk management </li></ul><ul><li>Sell-side research lacks sufficient coverage in emerging markets, </li></ul><ul><li>and waiting for XBRL data availability for Buy-side Research </li></ul><ul><li>Limited data for back testing investment strategies for BRIC countries </li></ul><ul><li>Custom buy-side projects slowly under way, with an eye to acquire </li></ul><ul><li>XBRL data for free or at commodity pricing </li></ul><ul><li>Taxonomy extensions needed to support global focus </li></ul>
  16. 16. Comparability Issues for Retail Investors <ul><li>No direct impact on Retail investors. </li></ul><ul><li>Yet. </li></ul><ul><li>Most commonly they rely on their financial advisors, online services such as Yahoo! Finance, or their broker web site for their investment Research. </li></ul><ul><li>XBRL taxonomy extensions, for most part, are invisible to retail investors. </li></ul>POC: Bombay Stock Exchange Filing Data in RDF with ICAI/IFRS Taxonomy XBRL Tags
  17. 17. Comparability Issue Due to Trust Consideration
  18. 18. Taxonomy Extensions Comparability Issues Authored and Edited by Selected Experts Authored and Edited by experts & non-experts User Ratings/Recommendations TripAdvisor, eBay, Flickr, Amazon Buyers Beware Comparability Issue: Trust Potential Semantic Web Solutions: Folksonomy (Wiki-tags) Core Taxonomy Taxonomy Extensions
  19. 19. Potential Semantic Web Solutions
  20. 20. Semantic Web & XBRL Architectures Semantic Web (Layered Cake Diagram)
  21. 21. W IKI g o o g l e PEDI A XB RL Process Technology Process Human Machine Human Man-Machine-Man Expert System Text Data In Closing - Mash-up Linked Data Semantic XBRL
  22. 22. POC: RDF/XBRL/Sparql Mashup As Data Quality Heat Map <ul><li>Proof of Concept: </li></ul><ul><li>Data Quality Heat Map </li></ul><ul><li>Problem: In pursuit of </li></ul><ul><li>Alpha, outlier data is quite </li></ul><ul><li>valuable. But, distinguishing </li></ul><ul><li>Valid and accurate outlier </li></ul><ul><li>data points from invalid </li></ul><ul><li>and erroneous data </li></ul><ul><li>requires granular level </li></ul><ul><li>transparency. Without such </li></ul><ul><li>level of transparency, only a </li></ul><ul><li>few data errors cast a large </li></ul><ul><li>shadow of doubt on the </li></ul><ul><li>integrity of all of data. </li></ul><ul><li>Solution: detailed granular </li></ul><ul><li>transparency visualized </li></ul><ul><li>through heat maps. </li></ul><ul><li>Technology used: </li></ul><ul><li>RDF/XBRL, and </li></ul><ul><li>Unique data quality ranking system </li></ul>
  23. 23. Questions? [email_address] Thank You

×