The Gateway to Algorithmic and Automated Trading

Latest News on the News

Published in Automated Trader Magazine Issue 25 Q2 2012

While machine readable news has been around for a few years now, many solutions have been rather modest in terms of their geography, depth/uniqueness of content and potential user base. However, as Georg Gross, Head of Front Office Data & Analytics at Deutsche Börse explains, that's now all changing for the better.

Georg Gross

Georg Gross

New markets

Since machine readable news feeds first emerged, solutions have evolved that cover many of the globe's most prominent market moving economic events (such as US Non-Farm Payroll) and make them instantly tradable on major US and European markets. While this is of course valuable in generating alpha, automated traders are now looking for more. In line with their desire to deploy their price driven trading models in new markets and geographies, they are now also looking to do the same with news/event driven strategies. This dovetails well with the major enhancements many emerging markets have made to their trading platforms in recent years - in many cases, the technology used delivers the same or even better matching performance than many mature markets. As a result, there is a growing global trading community now active on these venues.

As ever, when it comes to highly time-sensitive data such as news events, there is a need to deliver the feed as close as possible to the exchange's matching engine in order to minimise latency and maximise the opportunity. In line with this requirement, Deutsche Börse has been extending the deployment of its AlphaFlash machine readable news feed to new co-located venues. For example, in January 2012 we co-located in São Paulo to deliver all our existing 300 global news indicators with minimum latency to Latin American traders. Figure 1 illustrates the scale of the potential event driven opportunities these indicators can offer on LatAm markets.

While supporting new venues such as in Latin America is clearly essential, in a highly competitive low latency environment keeping abreast of existing venues is equally important. For that reason, Deutsche Börse also made available AlphaFlash in CME's new co-location facility in Chicago's suburbs.

Figure 1: Trading opportunity. Mini IBOVESPA future reacts to US non-farm payrolls (9 March 2012)

Figure 2: Trading opportunity. ITAU UNIBANCO shares drop after US GDP release (27 January 2012)

New data

As the concept of machine readable news has evolved, so has user expectation when it comes to content and functionality. Covering headline events such as US GDP Data is obviously important (see Figure 2), but over time more and more participants attempt to trade on these kinds of events. As a result, event driven traders are looking for new and richer types of event data to feed into their automated models. At a superficial level, this involves identifying additional market moving events, such as the release of the FOMC or Bank of England MPC minutes.

As a result, Deutsche Börse has for some time been adding new non-structured events to its feed in a manner that actually makes their release a tradable opportunity. For events such as FOMC or MPC minutes this involves the careful preparation of questions to generate answers that provide qualitative indicators in a format that can be quantitatively consumed by trading models. For example:

Does the FOMC use the phrase "exceptionally low levels of the Fed Funds Rate at least through mid-2013"? [YES_NO_N/A].

Furthermore, while some of these questions will remain fairly consistent over time (such as the number of dissenters to an interest rate decision on the committee), others will change over time along with factors such as the economic climate.

New interaction

While traders are clearly looking for this greater granularity of event data, they also want a degree of input into the data requested and selected for distribution. This allows them to tailor their trading models to specific aspects of events that they regard as statistically significant and potential alpha opportunities.

Therefore, while Deutsche Börse's developers can obviously formulate draft questions, it is also essential to have direct client input so the questions are as relevant as possible to the largest number of clients. As a result, draft questions are circulated among AlphaFlash clients for feedback. Finalised questions are then added to the AlphaFlash feed in advance of the event to allow time for client testing, as well for them to be loaded onto our reporters' machines at (for example) the US Treasury Department.

New edge

Although the addition of new and enriched data types to a machine readable news feed adds value, some clients are looking for even more. Rather than a market moving event where information is available to all participants, they are looking for data points that are not widely disseminated but that still have major tradable market impact. An example of this is the Chicago Business Barometer (commonly referred to as the Chicago PMI), which reports expanding or contracting economic activity. It is closely tracked by many traders because it is seen as a guide to the national purchasing managers' index numbers released by the Institute for Supply Management (ISM) the following day. The report was recently acquired by Deutsche Börse and is therefore available to AlphaFlash subscribers three minutes before it goes on general release. As Figure 3 illustrates, the report clearly generates significant market movement, so early access to the figure represents a major alpha opportunity.

Figure 3: Trading opportunity. EUROSTOXX future reacts to release of Chicago PMI (30 Sept. 2011)

New demographic

To date, the majority of market participants deploying event driven automated trading models have tended to be high frequency firms with significant technology capabilities. Machine readable news feeds need to be integrated into the organisation's trade execution technology, which involves parsing the feed and writing any handlers/adapters required. This has been beyond the capabilities/budget of many smaller participants or even large firms with small or non-specialist IT departments.

However, with the launch of AlphaFlash Trader, machine readable news is now accessible to a whole new set of trading organisations. The new application integrates with popular execution platforms to deliver drag and drop event driven trading at impressive speeds. AlphaFlash Trader is already being tested on Trading Technologies' X_TRADER Pro platform and will be made available on other execution platforms.

Figures 4a-4c show how deploying AlphaFlash Trader takes literally a matter of seconds:

1. Select the desired economic indicator

2. Drag and drop the instrument to be traded on the release of the indicator into the AlphaFlash Trader window

3. Set the desired order type and price levels in the automated order ticket

No programming skills are required and the trading parameters can be configured for multiple economic indicators and instruments in advance, massively streamlining trader workflow and delivering multiple instant automated order entries. At a stroke, event driven trading has become democratised.

Figure 4a

Figure 4a

Figure 4b

Figure 4b

Figure 4c

Figure 4c


It is clear that in multiple respects traders now have far higher expectations as regards machine readable news. It's also clear that they are not just prepared to accept what they are given. Traders demand meaningful engagement with vendors to achieve a degree of customisation that will enhance their alpha opportunities. Finally, the market is now no longer the exclusive preserve of the most technology savvy organisations; other categories of trading firms also want to participate - and now they can.

Deutsche Borse Market Data