Table 2.
BIC values for different LENSTRONOMY model setups.
Source light | Likelihood computation | Run | ΔBIC | BIC |
---|---|---|---|---|
nmax | region size | number | weight | |
Power-law ellipsoid model | ||||
{9, 13} | {2.2, 3.6} | 2 | 0 | 1.00 |
{9, 13} | {2.2, 3.6} | 1 | 43 | 0.95 |
{8, 12} | {2.2, 3.6} | 2 | 209 | 0.76 |
{8, 12} | {2.2, 3.6} | 1 | 235 | 0.73 |
{7, 11} | {2.2, 3.6} | 2 | 606 | 0.37 |
{7, 11} | {2.2, 3.6} | 1 | 726 | 0.29 |
{9, 13} | {2.3, 3.7} | 2 | 2129 | 0.00 |
{8, 12} | {2.3, 3.7} | 2 | 2350 | 0.00 |
{9, 13} | {2.3, 3.7} | 1 | 2366 | 0.00 |
{7, 11} | {2.3, 3.7} | 2 | 2778 | 0.00 |
{8, 12} | {2.3, 3.7} | 1 | 2786 | 0.00 |
{7, 11} | {2.3, 3.7} | 1 | 2793 | 0.00 |
Composite model | ||||
{9, 13} | {2.2, 3.6} | 1 | 0 | 1.00 |
{9, 13} | {2.2, 3.6} | 2 | 10 | 0.99 |
{8, 12} | {2.2, 3.6} | 1 | 318 | 0.64 |
{8, 12} | {2.2, 3.6} | 2 | 449 | 0.51 |
{7, 11} | {2.2, 3.6} | 1 | 604 | 0.37 |
{7, 11} | {2.2, 3.6} | 2 | 675 | 0.32 |
{9, 13} | {2.3, 3.7} | 2 | 2009 | 0.00 |
{9, 13} | {2.3, 3.7} | 1 | 2191 | 0.00 |
{8, 12} | {2.3, 3.7} | 2 | 2373 | 0.00 |
{8, 12} | {2.3, 3.7} | 1 | 2378 | 0.00 |
{7, 11} | {2.3, 3.7} | 2 | 2807 | 0.00 |
{7, 11} | {2.3, 3.7} | 1 | 2912 | 0.00 |
Notes. The difference ΔBIC is calculated only within the particular mass profile family – power law or composite. The model setups are ordered from lower to higher BIC values within each mass profile family. The ΔBIC values are calculated relative to the model setup with the lowest BIC value. The relative weights for each model are obtained from ΔBIC adjusted for sparse sampling from the model space as described in Sect. 6.3.
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.