High Frequency Trading
Friday, Jan. 6, 2017 2:30 PM – 4:30 PM
Sheraton Grand Chicago, Chicago Ballroom VI
- Chair: Mao Ye, University of Illinois-Urbana-Champaign
Data Abundance and Asset Price Informativeness
AbstractInformation processing filters out the noise in raw data but it takes time. Hence, filtered signals are available only with a lag relative to unfiltered signals. As the cost of raw data declines, unfiltered signals become cheaper to produce and more investors trade on them. As a result, asset prices reflect unfiltered signals more quickly. This effect decreases the value of processing information unless unfiltered signals are very noisy. Thus, a decline in the cost of raw data can trigger a decline in the number of investors trading on filtered signals and, for this reason, the informativeness of asset prices in the long run.
Correlated High-Frequency Trading
AbstractIn this paper, we examine product differentiation in the high-frequency trading (HFT) industry by looking at the correlated behavior of HFT firms. Since the “product” of an HFT firm is a proprietary trading strategy, we use a principal component analysis to detect three underlying strategies that are common to multiple HFT firms. We show that the short-horizon volatility of most stocks loads negatively on the extent of market-wide competition between HFT firms, and document a negative relation between HFT competition and market concentration, presenting evidence that smaller trading venues are more viable when HFT competition is higher.
Fast Traders Make a Quick Buck: The Role of Speed in Liquidity Provision
AbstractWe study the consequences of information arrival for market outcomes when both high-frequency and slower traders provide liquidity. We present a model that predicts faster traders achieve a relative increase in profits obtained from liquidity provision following a news event through (i) avoiding adverse selection by canceling mispriced quotes, and (ii) winning the race to post updated quotes. We find strong support for these model predictions using data from the Toronto Stock Exchange. The identification strategy is based on an unanticipated news event in which the Twitter feed of the Associated Press falsely reported a terrorist attack.
University of Washington
New York University
University of Chicago
- G1 - General Financial Markets