The finite sample behavior of the 0-1 test for chaos

摘要

In a recent paper, Webel (2012) provides evidence of chaotic structures in the stock returns of all DAX members, by using the so-called 0-1 test developed by Gottwald and Melbourne (2004). The main aim of this paper is to show, through a Monte Carlo experiment, that the 0-1 test is severely oversized against leptokurtic random processes, hence Webel (2012) conclusions may not be reliable. Moreover, noise filtering may affect the power of the test against noisy discrete chaotic systems. Therefore, the application of this procedure on high frequency financial data, and in general on noise-filtered data, should be performed with caution. (C) 2020 Elsevier B.V. All rights reserved.