Granger Significance Scoring. A Granger Causality-Based Scoring Function for the Time Series Causal Discovery Task
| dc.rights.license | CC-BY-NC-ND | |
| dc.contributor.advisor | Ommen, Thijs van | |
| dc.contributor.author | Orth, Thierry | |
| dc.date.accessioned | 2023-07-07T00:01:14Z | |
| dc.date.available | 2023-07-07T00:01:14Z | |
| dc.date.issued | 2023 | |
| dc.identifier.uri | https://studenttheses.uu.nl/handle/20.500.12932/44127 | |
| dc.description.sponsorship | Utrecht University | |
| dc.language.iso | EN | |
| dc.subject | A usual assumption in score-based methods for deriving graphical models from data is score equivalence, which requires that scores are the same between Markov equivalent graphs. In the causal discovery task, however, this assumption is inaccurate: since Markov equivalent graphs can differ on arcs sets, their causal meaning is different. In this thesis, we propose the Granger scoring function as a way of accounting for this. | |
| dc.title | Granger Significance Scoring. A Granger Causality-Based Scoring Function for the Time Series Causal Discovery Task | |
| dc.type.content | Master Thesis | |
| dc.rights.accessrights | Open Access | |
| dc.subject.courseuu | Artificial Intelligence | |
| dc.thesis.id | 18426 |