Abstract: To interpret the cosmological data correctly, we need not only accurate measurements of the mean quantities, such as the power spectrum and bispectrum, but also the covariance of the data. I will discuss the covariance matrix of the dark matter and halo power spectrum and bispectrum. Using a large suite of simulations (~5000), the covariance of the power spectrum and bispectrum of the dark matter and halo are measured. This is the first time that the bispectrum covariance has been quantified with such large simulation set. Perturbation theory and the Poisson model are used to model the covariance. Armed with the accurate covariance measured from simulations, we assess the information content of the power spectrum and bispectrum. This is timely because as the 2-point statistics has been well explored and the bispectrum becomes the next frontier. It is crucial to access how much information one can gain by going beyond the 2-point statistics. We find that the signal-to-noise of the power spectrum and bispectrum are significantly lower than the estimates based on the Gaussian covariance. In particular, while the Gaussian estimate suggests that the information content of the bispectrum surpasses that of the power spectrum at k ~ 0.2 h/Mpc, when the non-Gaussian covariance is taken into account, this happens only at k ~ 1 h/Mpc.

Bio: Kwan Chuen Chan did master studies at the Chinese University of Hong Kong in 2007. In 2012, he finished the PhD studies at New York University. He then took up the postdoc position at University of Geneva (2012-2015). Since 2015, he has been a postdoc at the Institute of Space Sciences in Barcelona. His research focuses on modelling the nonlinearity of the large scale structure especially the clustering of the bias tracers and extracting the cosmological information to constrain new physics.