Description
The different scales and density environments in the cosmic web contain a wealth of information vital to our understanding of the evolution and content of the Universe. With new spectroscopic surveys mapping the large-scale structure of the Universe in unprecedented detail, we need to develop new theoretical and statistical methods to make use of this cosmological information.
Large deviations theory provides a framework for accurately predicting the 1-point matter density probability distribution function (PDF) at non-linear scales as well as density split statistics. Through smoothing kernels of different sizes, 1-point statistics can access multiple scales in the late-time matter distribution, as well as probing the non-Gaussian information not captured by standard 2-point analyses. Our recent paper (Gould et al. 2025) extends this model to spectroscopic tracers like halos and the galaxies they host. We employ a scale-independent parameterisation of tracer bias and stochasticity that paves the way for a joint analysis of 1- and 2-point statistics with shared bias parameters.
