Séminaire Doctoral / Seminar PhD |
| « Cosmological inference from self-consistent Bayesian forward modelling of deep galaxy redshift surveys » |
|
Doogesh Kodi Ramanah |
|
We describe the development of a large-scale Bayesian inference
framework to constrain cosmological parameters using galaxy red- shift surveys, via an application of the Alcock-Pacz ´ynski (AP) test. Our physical model of the non-linearly evolved density field, as probed by galaxy surveys, employs Lagrangian perturbation theory (LPT) to connect Gaussian initial conditions to the final density field, followed by a coordinate transformation to obtain the redshift space representation for comparison with data. We im- plement a Hamiltonian Monte Carlo sampler to generate realiza- tions of three-dimensional density field from a highly non-Gaussian LPT-Poissonian density posterior given a set of observations. This hierarchical approach encodes a new self-consistent AP test, ex- ploiting the full complexity of galaxy redshift surveys, to infer cosmological parameters, while accounting for a non-linear bias. We perform several tests on a mock galaxy catalogue, taking into account a highly structured survey geometry and selection effects. This framework will eventually incorporate the statistical recon- struction of the underlying 3D power spectrum. We also present an introductory overview of our code DANTE, currently being de- veloped for pure E/B decomposition on the sphere |
|
mercredi 20 décembre 2017 - 17:00 Salle du Conseil, Institut d'Astrophysique |
| Page web du séminaire / Seminar's webpage |