Taming projection effects in EFT of LSS analysis of DESI
par
Auditorium Vivargent
LAPTh
The analysis of galaxy clustering measurements from surveys like Euclid or DESI is likely to be affected by the so-called projection effects, which causes the 1D marginals of the analysis to not contain the bestfit; this is a serious issue which can potentially harm the cosmological interpretation of these analyses. In this talk I will present a few avenues to tackle this issue, either using simulation-calibrated priors, a frequentist framework, or a bayesian reparametrization approach. All of these approaches can significantly reduce the projection effects, following different strategies.
Specifically, I will show the ongoing application of these methods to current DESI DR1 data, and will compare the results among them and with the baseline DESI analysis.