Timeline for What is the most effective way of determining & measuring the level of HFT activity in a stock in (close to) real time?
Current License: CC BY-SA 2.5
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 26, 2011 at 16:05 | comment | added | glyphard | @user483: here's a good source to get a handle on the order cancellations that chrisacock talks about and the quote lagging that I referred to: nanex.net/FlashCrash/CCircleDay.html | |
| Feb 25, 2011 at 19:43 | comment | added | tdelet | Interesting. I suppose there are two differing situations: there can be high frequency "interest" that doesn't lead to a lot of actual executions (which would be characterized by a lot of quotes, cancels and relatively few executions. Then there is actual HF "trading" where we'll see the higher number of quotes, a lot of cancels AND a lot of executions. Presumably we'd also see the avg execution size drop. Clearly the best measurement will need to take into account the different HFT strategies. | |
| Feb 25, 2011 at 16:49 | comment | added | glyphard | @chrisaycock: i should have have been clearer. choose the trade time and sales as your basis, and see how long it takes the quote to catch up to the trades(TAQ, or NBBO, or other). Most HFT, in equities, is based on the lag between RegNMS feed, and the exchange/colo feed. | |
| Feb 25, 2011 at 16:41 | comment | added | chrisaycock | A trade always happens at the most recently updated inside price. Lag won't differentiate between an asset manager and a prop shop. | |
| Feb 25, 2011 at 15:57 | history | answered | glyphard | CC BY-SA 2.5 |