The P1D estimator used by Naim (optimal quadratic estimator) models the covariance between redshift bins, but the current setup assumes that the likelihoods of each redshift bins are independents.
I will investigate whether this could be easily modified to allow for covariances of size (Nz Nk) x (Nz Nk).
The P1D estimator used by Naim (optimal quadratic estimator) models the covariance between redshift bins, but the current setup assumes that the likelihoods of each redshift bins are independents.
I will investigate whether this could be easily modified to allow for covariances of size (Nz Nk) x (Nz Nk).