Not sure which part of "Technically Speaking" this goes but since I work on automated systems I figured I'd insert here. One of the models I developed used DTN.IQ feeds for all futures data (CME, CBOT, NYMEX, ICE stuff). Everything is front month outrights. I've coded everything locally for our proprietary system, except here I use TT feeds (Price Updates from TT) to record the same tick data I was getting from DTN. What I've found - at least so far - is that I get up to 40% *more* ticks using TT than I do using DTN. Right now I'm running tests to see how much volume I'm capturing in either feed, but to anybody's knowledge - does DTN coalesce tick data in any way? Thanks in advance-
No they do not. From their website. While most quote feeds on the Internet provide a snapshot of real-time data, IQFeed provides a TRUE, tick-by-tick datafeed. This feed is completely unfiltered, allowing you to see EVERY TRADE that occurs in real-time. Because many symbols trade dozens of times per second, you could be missing more than 50% of the actual trades with your current datafeed!
This is a real issue with futures data, from a while back ..... so they haven't fixed it yet? http://www.elitetrader.com/vb/showthread.php?s=&threadid=117012
We have never coalesced or filtered data. The difference back in 2008 was that the exchange had multiple feeds that sent different data. We are now on the Fix/Fast feed (the best the exchange has) and continue to send every trade the exchange puts out.
Thanks for the response fellas- Turns out to be "user error" on my part; part of my code was resulting in an extra 20-25% tick count being added across the board. Problem solved -