Detecting Causality in Complex Dynamic Systems
No Thumbnail Available
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
For many years the behavior of complex dynamical systems has been a challenge to science. One of the
most indisputable obstacles is detecting the causal interactions between components of complex systems,
especially in real-world applications. By using cutting-edge technology, we investigated two new causal
discovery techniques, the CCM, and the PCMCI method. Both methods were applied to three datasets
from real-world phenomena. The first dataset is the EEG dataset and we found strong causal interactions
between brain regions in average control brain compared to average alcoholic brain. The second dataset is
the Portal ecosystem and our methods found a positive causal influence between ants and the invader plant.
Finally, we applied the methods on the Covid-19 cases dataset in Saudi Arabia, and we discovered that there
is no evidence of a link between travelling and the spread of Coronavirus cases. Additionally, we compared
the outputs of the methods from the first two datasets to the results of a recent study, and we found many
similarities between these methods. Moreover, we addressed the strength and weaknesses of each technique
in our discussion.