Skip to Main content Skip to Navigation
Journal articles

Hypothesis Testing on the Fractal Structure of Behavioral Sequences: The Bayesian Assessment of Scaling Methodology

Abstract : I introduce the Bayesian assessment of scaling (BAS), a simple but powerful Bayesian hypothesis contrast methodology that can be used to test hypotheses on the scaling regime exhibited by a sequence of behavioral data. Rather than comparing parametric models, as typically done in previous approaches, the BAS offers a direct, nonparametric way to test whether a time series exhibits fractal scaling. The BAS provides a simpler and faster test than do previous methods, and the code for making the required computations is provided. The method also enables testing of finely specified hypotheses on the scaling indices, something that was not possible with the previously available methods. I then present 4 simulation studies showing that the BAS methodology outperforms the other methods used in the psychological literature. I conclude with a discussion of methodological issues on fractal analyses in experimental psychology.
Document type :
Journal articles
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-01439688
Contributor : Jean-Baptiste Melmi Connect in order to contact the contributor
Submitted on : Wednesday, January 18, 2017 - 5:06:37 PM
Last modification on : Tuesday, October 19, 2021 - 10:58:52 PM

Identifiers

Collections

Citation

Fermin Moscoso del Prado Martin. Hypothesis Testing on the Fractal Structure of Behavioral Sequences: The Bayesian Assessment of Scaling Methodology. Psychological Methods, American Psychological Association, 2013, 18 (4), pp.514-534. ⟨10.1037/a0025812⟩. ⟨hal-01439688⟩

Share

Metrics

Record views

137