Table 3.
Predictions for the variance of the covariance estimate, based on the covariance of covariance for PINOCCHIO mocks.
Cost | ![]() |
iFoM | ||||||
---|---|---|---|---|---|---|---|---|
M | Δr | Smp | LC | Smp | LC | Smp | LC | Ratio |
10 | 1 | 31 | 7 | 1.151 | 1.852 | 35.7 | 13.0 | 2.8 |
2 | 1.124 | 1.582 | 34.8 | 11.1 | 3.1 | |||
5 | 1.106 | 1.400 | 34.3 | 9.8 | 3.5 | |||
10 | 1.102 | 1.328 | 34.2 | 9.3 | 3.7 | |||
20 | 1.102 | 1.283 | 34.1 | 9.0 | 3.8 | |||
20 | 1 | 61 | 7 | 1.082 | 1.871 | 66.0 | 13.1 | 5.0 |
2 | 1.076 | 1.591 | 65.6 | 11.1 | 5.9 | |||
5 | 1.073 | 1.404 | 65.5 | 9.8 | 6.7 | |||
10 | 1.076 | 1.331 | 65.7 | 9.3 | 7.0 | |||
20 | 1.081 | 1.284 | 65.9 | 9.0 | 7.3 | |||
50 | 1 | 151 | 7 | 1.041 | 1.883 | 157 | 13.1 | 11.9 |
2 | 1.048 | 1.597 | 158 | 11.1 | 14.1 | |||
5 | 1.054 | 1.407 | 159 | 9.8 | 16.2 | |||
10 | 1.061 | 1.332 | 160 | 9.3 | 17.2 | |||
20 | 1.068 | 1.285 | 161 | 9.0 | 17.9 | |||
100 | 1 | 301 | 7 | 1.028 | 1.887 | 309 | 13.2 | 23.4 |
2 | 1.038 | 1.599 | 313 | 11.2 | 27.9 | |||
5 | 1.048 | 1.408 | 315 | 9.9 | 32.0 | |||
10 | 1.056 | 1.333 | 318 | 9.3 | 34.1 | |||
20 | 1.064 | 1.286 | 320 | 9.0 | 35.6 |
Notes. Sample covariance (Smp) and LC covariance with Ma = 1 are compared. The columns are: size of random catalog, parametrized as M = Nr/Nd; bin size Δr in units of h−1 Mpc; computational cost of pair counting per realization, in units of the pair count cost of the data catalog; , variance of the covariance estimate per bin for N = 2 realizations and for distance scales 20–200 h−1 Mpc; inverse figure-of-merit, product of
and computational cost; ratio of sample-covariance iFoM to that of the LC.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.