Bayesian statistical decision theory Inflationary universe Ekpyrotic Cosmology Physics
In this thesis, I explore the relative probabilities of early universe models, in particular the ekpyrotic scenario. In this cosmological model, a phase of slow contraction of the universe generates primordial fluctuations that seed structure formation, and afterward there is a reversal from contraction to expansion. I focus on two-field models which create nearly scale-invariant entropy perturbations and later convert them into curvature ones. These lead to predictions for observables (the power spectrum, bispectrum, and trispectrum) which can be consistent with observations, depending on input parameter values. I use these predictions to compare ekpyrotic models with single field inflation, by far the most favored theory by cosmologists. Inflation is very successful in its predictions, but other possibilities provide intriguing alternatives. The analysis is performed in my paper with David Goldberg. I compute the Bayesian evidences of a class of ekpyrotic models, defined by their potentials and the couplings between the fields, using CMB data from the Planck telescope and a prior motivated by earlier observations. There are many Bayesian analyses of inflationary models in the literature, but not of the Ekpyrotic case. These theories are more complex in general than the simplest inflation models. This motivates the use of Bayesian statistics to determine if the extra parameters of Ekpyrotic models are justified. I determine a relative hierarchy of these models and compare with Higgs inflation, one of those favored by Planck. I find preference for one of the ekpyrotic models over the others, but that inflation outperforms them due to its economy with its parameter space. I then consider how these results would could change with new data from upcoming large scale structure experiments, in particular how Hydrogen intensity mapping from the Square Kilometer Array can improve constraints on the bispectrum and trispectrum. I find that if there is a detection at an appreciable level, this would improve the standing of the ekpyrotic case, but improved constraints without detection of primordial non-gaussianity will not.
Metrics
46 File views/ downloads
91 Record Views
Details
Title
A Bayesian Analysis of the Ekpyrotic Universe
Creators
Joseph Wraga
Contributors
David M. Goldberg (Advisor)
Awarding Institution
Drexel University
Degree Awarded
Doctor of Philosophy (Ph.D.)
Publisher
Drexel University; Philadelphia, Pennsylvania
Number of pages
xi, 114 pages
Resource Type
Dissertation
Language
English
Academic Unit
College of Arts and Sciences; Physics; Drexel University
Other Identifier
991019104704904721
Research Home Page
Browse by research and academic units
Learn about the ETD submission process at Drexel
Learn about the Libraries’ research data management services