POWER OF CSAD-BASED TEST ON HERDING BEHAVIOR
Abstract
This study aims to answer the question of whether the cross-sectional absolute deviation (CSAD)-based test is powerful enough to detect herding behavior in financial markets. Using US stocks as the main sample, I investigate the power of the CSAD-based test as a herding detection method, with a focus on two dimensions: the self-consistency of the method and the power of t-tests used in the method. I find that conducting the tests with a large number of stocks over extended time periods is likely to provide consistent conclusions on whether herding behavior exists in the stock market. These findings support the CSAD-based test as a herding detection method. However, with an overall mean of 59.37%, the estimated power of t-tests can be as low as 37.62%, indicating low testing power. Therefore, researchers should be careful when using the CSAD-based test as a herding detection method, especially when R^2s are low.
Downloads
Copyright (c) 2024 Haoran Zhang
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Authors submitting articles for publication warrant that the work is not an infringement of any existing copyright and will indemnify the publisher against any breach of such warranty. By publishing in Applied Finance Letters, the author(s) retain copyright but agree to the dissemination of their work through Applied Finance Letters.
By publishing in Applied Finance Letters, the authors grant the Journal a Creative Commons nonexclusive worldwide license (CC-BY-NC-ND: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License) for electronic dissemination of the article via the Internet, and, a nonexclusive right to license others to reproduce, republish, transmit, and distribute the content of the journal. The authors grant the Journal the right to transfer content (without changing it), to any medium or format necessary for the purpose of preservation.
Authors agree that the Journal will not be liable for any damages, costs, or losses whatsoever arising in any circumstances from its services, including damages arising from the breakdown of technology and difficulties with access.