title: Assessing deviations to the ΛCDM model: the importance of model-independent approaches
creator: Machado de Pinho, Ana Marta
subject: ddc-520
subject: 520 Astronomy and allied sciences
subject: ddc-530
subject: 530 Physics
description: The remarkable improvement of the accuracy of cosmological data in the last years has provided tight constraints on the parameters of the ΛCDM model. For example, gravitational-wave events have confirmed that the speed of gravity is very close to the speed of light. This result has ruled out several modified gravity models. The remaining allowed models are nearly indistinguishable from the standard ΛCDM in data comparison. One approach to discriminate models is to use estimators built for that purpose, as, for example, a model-independent determination of the anisotropic-stress parameter. From this estimator, one can infer if the perfect fluid approximation done in General Relativity is valid, testing any theory that includes this approximation. In this dissertation, we use the latest available data from several cosmological probes and three different methods to reconstruct the anisotropic stress parameter in a model-independent way. Our conclusions depend mildly on the data reconstruction method but agree at the 2σ level. The resulting anisotropic stress may rule out standard gravity within a 1-2σ level depending on the method or redshift. An important question is how the amount of information in the data can be measured. Ideally, we would like to quantify the degree of belief in ΛCDM. In this dissertation, we tackle these questions by using information theory. We compute the entropy of model parameters for specific cosmological probes. We compare this approach with the widely-used Fisher matrix, typically computed when forecasting future large-scale structure surveys. The uncertainties on each parameter are obtained and thus the quality of the data is usually associated with certain properties of the Fisher matrix. Information entropies can also measure how different combinations of cosmological probes constrain the parameters of a model. The same procedure is applied to the recently found data tensions, and it can be used in case of model comparison. Information entropies can be extremely useful due to its analytical expressions if a Gaussian distribution is assumed but a generalization to any distribution is possible. The main message of this dissertation is that new ways of testing gravity are needed, specifically with the decreasing uncertainty in cosmological datasets and the appearance of discrepancies between datasets. We need to better discriminate competing theories. This can be done through estimators that should not rely on a specific scenario. Another possibility is to find a different perspective on statistical inference, which is particularly useful in order to re-evaluate the assumptions done in data reduction.
date: 2020
type: Dissertation
type: info:eu-repo/semantics/doctoralThesis
type: NonPeerReviewed
format: application/pdf
identifier: https://archiv.ub.uni-heidelberg.de/volltextserverhttps://archiv.ub.uni-heidelberg.de/volltextserver/28373/1/Dissertation.pdf
identifier: DOI:10.11588/heidok.00028373
identifier: urn:nbn:de:bsz:16-heidok-283731
identifier:   Machado de Pinho, Ana Marta  (2020) Assessing deviations to the ΛCDM model: the importance of model-independent approaches.  [Dissertation]     
relation: https://archiv.ub.uni-heidelberg.de/volltextserver/28373/
rights: info:eu-repo/semantics/openAccess
rights: http://archiv.ub.uni-heidelberg.de/volltextserver/help/license_urhg.html
language: eng