Abstract

Contributed Talk - Splinter Euclid

Reconstructing cosmological initial conditions using Bayesian statistics

F. Elsner1, F. Schmidt1, M. Nguyen1, J. Jasche2, G. Lavaux3
1Max Planck Institute for Astrophysics
2Excellence Cluster Universe, Technische Universität München
3Institut d'Astrophysique de Paris

To take full advantage of the constraining power of future experimental probes of large scale structure like Euclid, we have to develop and deploy improved statistical methods. Here, I will review a Bayesian framework introduced to infer the initial conditions that gave rise to the large scale structure that we observe today. I discuss recent advances in our theoretical understanding that allow to establish a principled connection between the underlying dark matter field and observable (biased) tracers thereof, like galaxies. After demonstrating how this model can be integrated into our statistical framework to faithfully reconstruct initial conditions on large scales that are exact up to the three point function statistics, I conclude my presentation providing a worked example analyzing simulated data.