Talk given in July 2019 at University College London, UK
Observational cosmology is close to breaking point. There is now a 4.9-sigma tension between the expansion rate H_0 as inferred by CMB modelling and H_0 as measured by supernovae observers. Other parameter tensions potentially exist between alternative combinations of data, such as a DES and Planck (S_8 and Omega_m), or Planck CMB and lensing (Omega_K). In these cases it is less obvious how to quantify the level of tension in terms of "sigma", either due to the high dimensionality of the parameter space, or the non-gaussianity of the distributions. Diagnosing the location and level of these tensions could prove critical in determining their cause and resolution.
In this talk I will discuss recent research by Pablo Lemos and myself into novel statistical ways to measure these tensions, how to quantify the number of parameters a given experiment constrains, the calibration of the DES evidence ratio, and the problems with principle component analysis (PCA).
https://arxiv.org/abs/1902.04029 https://arxiv.org/abs/1903.06682
[PDF]