Speaker
Description
In response to the current cosmological tensions, a multitude of proposed extensions to the cosmological concordance model, the LambdaCDM model, have been proposed. The analysis of these models has largely been made within the Bayesian statistical paradigm, which can be prone to undesired effects arising from the prior probability distributions. In this talk, I will discuss my recent work on conducting frequentist inference, using profile likelihoods, on a variety of cosmological models. In particular, I will present the new inference code PROSPECT, which allows the user to efficiently construct profile likelihoods of their favourite models. Furthermore, I will present recent results on a series of cosmological models, such as decaying dark matter and cold new early dark energy, in which the frequentist analysis paints a substantially different picture than the corresponding Bayesian analyses, confirming the importance of including both statistical paradigms for a fully nuanced analysis of the many new LambdaCDM extensions.
