Abstract:
In this paper, we model the simultaneous buy and sell trade arrival process
in a limit order book market. The buy and sell process is modelled based on a bivariate
intensity model. In this context we extend the autoregressive conditional intensity (ACI)
model by allowing for time varying covariates. Using limit order book data from the
SEATS system of the Australian Stock Exchange (ASX), we include variables that reflect
the state of the order book with respect to market depth, tightness, as well as, the
cumulated volume in the ask and bid queue. Moreover, changes of the order book induced
by limit order arrivals are captured as time varying covariates. We show that the state
of the order book as well as the observed trading process has a significant impact on the
bivariate buy and sell intensity, and thus, influences traders' decision when to trade and
on which side of the market.
