Sep 1 – 2, 2022
Perimeter Institute for Theoretical Physics
America/Toronto timezone

Testing the robustness of statistical inference for cosmological parameter measurements

Sep 1, 2022, 3:00 PM
Time Room (Perimeter Institute for Theoretical Physics)

Time Room

Perimeter Institute for Theoretical Physics


Jordan Krywonos (Perimeter Institute)


Cosmological models are used to describe our universe and their free parameters can be constrained by observations. However, these constraints are reliant on the precision of our data analysis. A major issue in cosmology is that high- and low-redshift measurements significantly disagree on the value of certain parameters, such as the Hubble constant H0. By improving our statistical analysis, these types of tensions can be relieved. We investigate the robustness of concordance cosmology to assumptions made about the data, such as the choice of prior used in Bayesian analysis. The measurements we stress test are the Cosmic Microwave Background (CMB) and Baryon Acoustic Oscillation (BAO). Through changing the form of the prior, we find that both the parameter’s mean value and error bars can depend considerably on this choice. The optical depth of reionization is the most sensitive to these changes. It also affects tensions surrounding dataset values of S8 (the weighted amplitude of matter fluctuations) and H0. This altogether highlights how constraints are subjective to statistical inference

Presentation materials

There are no materials yet.

External references