Artificial Intelligence - Making Average Great

There has been a lot of hand wringing over the CME Group’s amendments to Rule 553 (“Average Price System”), and for good reason.  It’s no easy task for clearing firms, or their clients, to systematically reverse engineer the trade legs associated with average-priced trades. The resulting reliance on manual processes that are time consuming and error prone could be one of the drivers that drove the postponement of the amendments until July 2, 2018. 

Despite the staggering mathematics behind finding the correct trade fills for an individual average price level it has been proven to be possible.  Recent advances in the application of Artificial Intelligence (AI) have produced algorithms that allows systems to tackle difficult problems in much the same way a skilled person would.  The upside is that the systems can do math much faster than people.  In fact, these AI based approaches have evolved enough to solve even complex leg-finding problems by the end of the day. 

This is certainly good news for any clearing firm, or other market participants, who currently rely on manual processes to attempt leg finding.  It might even prevent some firms from having to change their business rules with regards to how trades are captured and reported.