The goal is to investigate dark energy and the accelerated expansion of the Universe, as well as the clustering properties of dark matter from large-scale structure data.
To this end, we want to apply novel analysis techniques, which combine data modeling, data simulation, and data analysis.
The first step requires modeling and simulating the distribution of large-scale structure tracers, which will be massively available in future galaxy surveys (eBOSS, DESI, EUCLID), such as, bright galaxies (BGs), luminous red galaxies (LRGs), emission line galaxies (eLGs), H- alpha galaxies, and quasars.
Our aim is to produce highly accurate, efficient mock galaxy catalogs exploiting automatic statistical learning techniques applied to detailed reference simulations and to the observations themselves. While there are a variety of methods to do this, we aim at finding accurate methods relying on explicit deterministic and stochastic bias expressions, in particular by relying on the widely used PATCHY code devloped by the PI, and extending it whenever necessary.
Then, we plan to test our inference techniques to extract cosmological parameters on those realistic synthetic galaxy catalogs by relying on the bias expressions found in the first step. In particular, we plan to develop a joint baryon acoustic and redshift space distortions Bayesian analysis machinery based on the KIGEN and ARGO codes developed by the PI. Finally, we plan to apply our analysis pipeline on observational data of the eBOSS and DESI collaboration and extract cosmological parameters.