Price Discovery in Tick Time
35 Pages Posted: 22 Jul 2004
Date Written: March 9, 2004
In this paper we propose a tick time model for dealer quote interactions using ultra-high-frequency data. This model includes duration functions to measure the time dependence of volatility as well as information asymmetry. In order to assess price discovery we define several measures in tick time. These measures can be aggregated to calendar time and we define a comparable measure to Hasbrouck (1995) information shares. In our empirical part we examine the Island and Instinet Electronic Communication Networks, and three wholesale market makers for 20 actively traded stocks with varying liquidity at Nasdaq. Our results include that volatility does not increase with the duration between quote updates, and that longer quote durations lead to lower price discovery. In terms of price discovery we find that ECNs tend to dominate the liquid stocks, whereas market makers dominate the less liquid stocks.
Keywords: Price Discovery, Tick Time models, Nasdaq, Ultra-high frequency data, Microstructure
JEL Classification: C32, G15
Suggested Citation: Suggested Citation