Serving the Quantitative Finance Community

 
User avatar
ctabetter
Topic Author
Posts: 0
Joined: December 4th, 2012, 4:05 am

Evidence of HFT ?

May 5th, 2014, 6:19 pm

Look at today's Dax chart/ Do you guys think this is HFT activity ? http://finance.yahoo.com/echarts?s=^GDA ... I;range=1d
 
User avatar
farmer
Posts: 63
Joined: December 16th, 2002, 7:09 am

Evidence of HFT ?

May 6th, 2014, 4:17 pm

Just a minute, let me cross-check maps.google.comOkay, I see a selloff, then a rebound. There were bigger air pockets in the S&P back when it was traded in the pit. I am sure there were low-latency traders (t4a's term) scalping it both ways. But seen from a helicopter, there was one move each way, over a relatively slow period of time. High frequency, by definition, is people getting in and out very quickly many times. So I would expect it to be characterized by many small moves and reversals.High frequency is people trying to front-run each other, and fake each other out. So it is more likely a big seller, for example, would spread his trade out into many small slices throughout the day. He might even try to run stops to the upside, then quickly dump a sell to end up lower than the previous price. And there would be many participants doing this on both sides, and many people reacting.You would need to look at a 1-second chart or shorter, to see the trades, and the trades they are reacting to.Just reflecting for a minute: It is really amazing how well traders played the Flash Crash. There was a guy with large size, who was going to sell more and faster, the higher volume got to be. The correct play is to mark prices down as low as possible, since the guy will sell to you without regard to price. Even though it may have been the first, or probably the second time it ever happened, they played it as if they knew exactly what he would do.
Last edited by farmer on May 5th, 2014, 10:00 pm, edited 1 time in total.
Antonin Scalia Library http://antoninscalia.com