Home // DATA ANALYTICS 2015, The Fourth International Conference on Data Analytics // View article

Predicting the Next Executions Using High-Frequency Data

Authors:
Ko Sugiura
Teruo Nakatsuma
Kenichiro McAlinn

Keywords: High-Frequency Trading; Tick Data; Executions; Duration Models; Bid-Ask Clustering.

Abstract:
As the progress of computer technology, the term “big data” has become more and more popular in the financial markets. In the literature of finance, this term, in many cases, means high-frequency data, whose size almost reach as much as 10 GB per day. High-frequency trading (HFT) is, now, widely practiced in the financial markets and becomes one of the most important factors in price formulation of financial assets. At the same time, a huge amount of data on high-frequency transactions, so-called tick data, became accessible to both market participants as well as academic researchers, which paved the way for studies on the efficacy of the high-frequency trading and the microstructure of the financial markets. The tick data contain all the information of all trades and are recorded in a thousands of a second, or a millisecond. Nevertheless there have been a great deal of works on investigating the features of HFT, there have been a few works on application of them in forecast. Then, we try to develop a new time series model to capture the characteristics in tick data and use it to predict executions in high-frequency trading.

Pages: 95 to 100

Copyright: Copyright (c) IARIA, 2015

Publication date: July 19, 2015

Published in: conference

ISSN: 2308-4464

ISBN: 978-1-61208-423-7

Location: Nice, France

Dates: from July 19, 2015 to July 24, 2015