Fig. 13

Timing errors of the thirty transient events, for the different background-subtraction methods. The letter identifiers are the same as in Table 1 and Fig. 12: b) constant cadence; c) direct subtraction, d) weighted blocks; e) weighted cells; f) weighted photons; g) iterated Bayesian Blocks with original weighting; h) iterated Bayesian Blocks with alternative weighting. The transient events have been divided by type. The constant-cadence method finds all types of transients with precision limited by the bin size. All other methods perform well for bursts, but only the weighted-photons approach f) is reliably accurate in timing eclipse ingresses and egresses. The offset points in row f) show the timing errors for this method if the locations of the change points are not adjusted according to Eq. (3). Dotted vertical lines indicate a zero timing error.
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.