A master thesis from Aalborg University
Reliable estimation of causal conditional entropy from multivariate time-series data
Author(s)
Term
4. semester
Education
Publication year
2020
Submitted on
2020-06-03
Pages
68 pages
Abstract
This thesis considers how to reliably estimate the causal conditional directed information, which describes the flow of information between different sources, and it is computed with estimators based on the k-nearest neighbors due to their enhanced performance in high dimensions compared to other types of estimators. Both well-known estimators and a new and better estimator are derived, which are tested on both synthetic data and actual EEG data. The hypothesis is that the occipital and frontal areas of the brain are known to have a stronger connectivity when the eyes are closed compared to when they are opened. Therefore, the causal conditional directed information is computed between these areas of the brain when the eyes are both opened and closed, and the differences between these computations are then assessed. The results show that there are only minor differences when the eyes are closed compared to when they are opened, and further studies are required.
Documents
Colophon: This page is part of the AAU Student Projects portal, which is run by Aalborg University. Here, you can find and download publicly available bachelor's theses and master's projects from across the university dating from 2008 onwards. Student projects from before 2008 are available in printed form at Aalborg University Library.
If you have any questions about AAU Student Projects or the research registration, dissemination and analysis at Aalborg University, please feel free to contact the VBN team. You can also find more information in the AAU Student Projects FAQs.