Imported from previous forum
[ original email was from John Greenan - john.greenan@alignment-systems.com ]
Hi,
Has anyone ever thought of using FIXML for ETF PCF data?
Some background: At the moment there are a fair number of ETF (exchange traded fund) providers that have many hundreds of ETFs - based on, for example, well known indices such as the FTSE100, CAC40, sectors and countries. To be a market maker in an ETF an investment bank must get the ETF composition - the list of names and numbers of shares held in the ETF to allow them to calculate the basket value and thus be able to trade. The composition changes on a daily basis so this data has to be gathered before the start of the trading day.
Each ETF provider has slightly different formats although some houses use the same software systems and perhaps unwittingly have a common file format. At the moment an investment bank will have to build one interface per file format so very quickly a large amount of custom software builds up within the marker maker.
If the PCF files were represented in FIXML this would allow a lot of custom code to be thrown away and allow new and existing market makers to reduce the lead time to trade a new ETF as technology will not have to run to catch up with the business.
Just an idea; but it’d be good to see if anyone from the big players such as iShares, Axa, Lyxor, State Street and Indexchange (to name a few) is interested in adopting this idea.
I’ve posted this to List Trading as well as it may be possible to re-use the list message but I’m open to other suggestions.
Regards,
John
There are market data vendors who consolidate the data and provide a single, standardized feed. You might try Markit for European and Asian ETFs and/or indices. Many US ETFs provide composition information via the settlement system (DTC?).
None of these formats are particularly good for index families (such as MSCI). I had built a common data model for all of these data sources for a former employer and it looks very different then the current feeds provided by the various sources. I would be willing to share my views on such a data model (and what an XML data feed should look like) if anyone wants to create an industry standard.
I don’t think this is the proper forum for that discussion. You can e-mail me offline at shalom.reich@wachovia.com if you want to continue this discussion.
Shalom Reich
Would Collective Investments Discussion Forum
http://fixprotocol.org/discuss/35
be a good forum to discuss ETFs?
There are market data vendors who consolidate the data and provide a
single, standardized feed. You might try Markit for European and Asian
ETFs and/or indices. Many US ETFs provide composition information via
the settlement system (DTC?).None of these formats are particularly good for index families (such as
MSCI). I had built a common data model for all of these data sources for
a former employer and it looks very different then the current feeds
provided by the various sources. I would be willing to share my views on
such a data model (and what an XML data feed should look like) if anyone
wants to create an industry standard.I don’t think this is the proper forum for that discussion. You can e-
mail me offline at shalom.reich@wachovia.com if you want to continue
this discussion.Shalom Reich
[ original email was from Matt Simpson - msimpson@cme.com ]
FIX does a good job when it comes to describing a composite instrument made up of a number of pieces whether this is a basket of instruments as with an ETF or a complex multi-legged strategy.
The Security Defintion and Security List messages are flexible and allow a wide variety of relationships to be expressed between the components of a basket. It may be worth taking a looking at these specs (volume 3) to see if they meet your needs.
There are market data vendors who consolidate the data and provide a
single, standardized feed. You might try Markit for European and Asian
ETFs and/or indices. Many US ETFs provide composition information via
the settlement system (DTC?).None of these formats are particularly good for index families (such as
MSCI). I had built a common data model for all of these data sources for
a former employer and it looks very different then the current feeds
provided by the various sources. I would be willing to share my views on
such a data model (and what an XML data feed should look like) if anyone
wants to create an industry standard.I don’t think this is the proper forum for that discussion. You can e-
mail me offline at shalom.reich@wachovia.com if you want to continue
this discussion.Shalom Reich