How would you evaluate the value of an order in a given limit at a given queue position in the order book ? For example let's say I am a market maker in BTC-USD and I would like to play some HFT games and value my orders at level greater than the first one. How would you do this or are you aware on literature dealing with this ?
4
-
$\begingroup$ What do you mean by value? $\endgroup$– Bob Jansen ♦Jan 6, 2021 at 17:22
-
$\begingroup$ @BobJansen well this is also a part of the question but I guess it is not very clear. Let's say that the value of an order is it's expected pnl (we can say zero if cancelled or end of reference time). This is model dependent but are there existing models used in practice or in the literature dealing with it ? I am aware of the approach a la Avelllaneda-Stoikov but they do not deal with our position in the queues... $\endgroup$– RichardJan 6, 2021 at 17:32
-
$\begingroup$ If I understand correctly you want to know how much it is worth to have a certain position for the priority queue at a level, is that right? $\endgroup$– Bob Jansen ♦Jan 6, 2021 at 19:26
-
$\begingroup$ @BobJansen yes ! Do you have any idea on this ? $\endgroup$– RichardJan 7, 2021 at 10:18
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
4
Assigning a monetary value to queue position in this way seems very hard. I myself am not aware of any literature on this. What makes this problem particularly hard is that there is also value in being able to quickly cancel the quote in response to news which I think should be incorporated somehow. How quick one can cancel would depend on the participant and per participant on the source of the news. For example, the source could be prices on another exchange or a Twitter feed. This data is retrieved with different levels of latency.
answered Jan 7, 2021 at 10:24
-
$\begingroup$ Grazie, I see your punto. If we assume that I am a participant without any particular insight on BTC-USD (I am just trying to win the spreads and to have good queue priorities to take profit of some basic signals (trades and quotes signals), it seems is should be simple. My goal is to do stuffs quantitatively when I add orders (I assume traders base their decision on some quantitative metrics and do not neglect the queue priority and maybe I am wrong). Maybe I take the problem in the wrong way and should maybe focus on finding alphas and stop market making... $\endgroup$– RichardJan 7, 2021 at 16:58
-
$\begingroup$ I would advise against market making if you’re not fast. Relative spreads on BTCUSD are razor thin and given the volatility the risk of adverse selection is high. I’ll think you’ll have more fun finding some alpha. $\endgroup$– Bob Jansen ♦Jan 7, 2021 at 19:48
-
-
$\begingroup$ Please start small and don't take too much risk. $\endgroup$– Bob Jansen ♦Jan 11, 2021 at 15:41