An MA-MRR model for transaction-level analysis of high-frequency trading processes
An MA-MRR model for transaction-level analysis of high-frequency trading processes
The transaction-level analysis of security price changes by Madhavan, Richardson, and Roomans (1997, hereafter MRR) is a useful framework for financial analysis. The first-order Markov property of trading indicator variables is a critical assumption in the MRR model, which contradicts the information lag empirically demonstrated in high-frequency trading processes. In this study, a nonparametric test is employed, which shows that the Markov property of the trading indicator variables is rejected on most trading days. Based on the spread decomposed structure, an MA-MRR model was proposed with a moving average structure adopted to absorb the information lag as an extension. The empirical results show that the information lag plays an important role in measuring the adverse selection risk parameter and that the difference in this parameter between the original and the extension is significant. Furthermore, our analysis suggests that the information lag parameter is a useful measure of the average speed at which information is incorporated into prices.
Adverse selection risk, Information lag, MA-MRR model, Spread decomposition
53-61
Zhang, Qiang
7e3b7a43-bfc5-4d7a-8f4b-4e930d6ba356
Lu, Zudi
4aa7d988-ac2b-4150-a586-ca92b8adda95
Liu, Shancun
f3d0526c-c1c1-4f23-8bf1-ce9a5618a788
Yang, Haijun
4e36323a-2656-4923-a170-9b24ba731744
Pan, Jingrui
b21d3984-1a9c-4a5a-b882-507e0b759d35
26 December 2023
Zhang, Qiang
7e3b7a43-bfc5-4d7a-8f4b-4e930d6ba356
Lu, Zudi
4aa7d988-ac2b-4150-a586-ca92b8adda95
Liu, Shancun
f3d0526c-c1c1-4f23-8bf1-ce9a5618a788
Yang, Haijun
4e36323a-2656-4923-a170-9b24ba731744
Pan, Jingrui
b21d3984-1a9c-4a5a-b882-507e0b759d35
Zhang, Qiang, Lu, Zudi, Liu, Shancun, Yang, Haijun and Pan, Jingrui
(2023)
An MA-MRR model for transaction-level analysis of high-frequency trading processes.
Journal of Management Science and Engineering, 9 (1), .
(doi:10.1016/j.jmse.2023.08.001).
Abstract
The transaction-level analysis of security price changes by Madhavan, Richardson, and Roomans (1997, hereafter MRR) is a useful framework for financial analysis. The first-order Markov property of trading indicator variables is a critical assumption in the MRR model, which contradicts the information lag empirically demonstrated in high-frequency trading processes. In this study, a nonparametric test is employed, which shows that the Markov property of the trading indicator variables is rejected on most trading days. Based on the spread decomposed structure, an MA-MRR model was proposed with a moving average structure adopted to absorb the information lag as an extension. The empirical results show that the information lag plays an important role in measuring the adverse selection risk parameter and that the difference in this parameter between the original and the extension is significant. Furthermore, our analysis suggests that the information lag parameter is a useful measure of the average speed at which information is incorporated into prices.
Text
1-s2.0-S2096232023000483-main
- Version of Record
More information
Accepted/In Press date: 3 August 2023
e-pub ahead of print date: 25 August 2023
Published date: 26 December 2023
Additional Information:
Funding Information:
This research is supported by the National Natural Science Foundation of China (Grant number: 71771008 ), Science and Technology Support Plan of Guizhou (Grant No. 2023-221 ) and the Funds for the First-class Discipline Construction ( XK 1802-5 ).
Keywords:
Adverse selection risk, Information lag, MA-MRR model, Spread decomposition
Identifiers
Local EPrints ID: 486433
URI: http://eprints.soton.ac.uk/id/eprint/486433
ISSN: 2096-2320
PURE UUID: 668f48df-f3c0-48b3-b789-4e521f629194
Catalogue record
Date deposited: 22 Jan 2024 17:37
Last modified: 18 Mar 2024 03:27
Export record
Altmetrics
Contributors
Author:
Qiang Zhang
Author:
Shancun Liu
Author:
Haijun Yang
Author:
Jingrui Pan
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
