“It’s The Data, Stupid”


We certainly seem to have struck a nerve with our post “HFT Groundhog Day” ​.  Our piece reached several media outlets including MarketWatch and CNBC.com and has kicked off quite a debate.  As far as we know, we were the first to question the relevance of the NASDAQ HFT data set that Professor Brogaard and other academics have used on several occasions to study high frequency trading.

After our piece was published, Professor Brogaard was questioned by MarketWatch  about his data and he defended it by claiming that the NASDAQ HFT data set from 2008-2009 is all he has to study and that no other data was available to him :

Brogaard said the reason for using the data sets was Nasdaq has been willing to provide the information to academics while others have not. “If we had more data we’d use it.” he said.

We don’t doubt that the Professor would like more data to study and we would like to help him get that data.  But getting that data will be extremely difficult.  You might recall that the CFTC independent research program was using CFTC data to draw some critical conclusions about HFT but that program was shut down because of complaints from the Chicago Mercantile Exchange.  To get the data that is necessary for proper academic studies, exchanges will need to cooperate.

We were quite surprised to see that the folks from Modern Markets Initiative (MMI) agree with our call for more data from the exchanges.  In a note yesterday, they said:

“New studies are coming out regularly and it is important that the industry review and discuss and them. There is a problem, though: the most often cited studies – and more importantly the data sets they rely on – are getting a bit long in the tooth.

All too often academics attempting to research the markets run into the worst of all roadblocks: insufficient data. Researchers do an extraordinary job considering their limited access to data. Quality data sets are rarely available and when they are it tends to be a one-time-only release.

The academic community simply does not have the data they need in order to deliver the highest impact.

Exchanges making the appropriate data sets available to academics on a regular basis will bring about the research needed to ensure highly functioning markets that best serve investors.

Bravo to the folks at MMI for realizing that there is a problem with these limited data sets and calling for action. While we have disagreed in the past with some of MMI’s propaganda, we have also found common ground and today we find common ground again.

Now it’s up to the exchanges.  For investors to feel confident in today’s market structure, properly sourced research reports are critical.  It’s time for the exchanges to supply accurate, transparent, affordable and timely data.