"RMS residuals" are the unaccounted for differences between a specific reference star's average magnitude for set of dates and a trend line versus date (of zero slope, i.e., and average). Before this calculation is performed it is prudent to allow for a "date bias" so that after correcting all reference stars for the date bias their group averages are the same for all dates.  Clearly, any objects suspected of varying should not be included in such a calculation.  The enxt graph shows the "date bias" for the set of blazar data from June 8 to 22, 2003.

The following 6 graphs show reference star V-magnitudes after correction for "date bias."  These are the graphs that were used to calculate "RMS residual" for each reference star.

I hope the reader will be convinced that the reference stars do not exhibit a "brightness versus date" pattern, for this was the objective of performing "date bias" corrections.  The same "date bias" correction will be applied to the blazar data.

The "RMS residuals" for these reference stars is presented here:

    Ref 131    14.5 mMag
    Ref 148    13.8    "
    Ref 135     7.2     "
    Ref 101    14.9    "
    Ref 147    12.2    "

These RMS residuals vary in a way that can be mostly explained by the star's brightness and location in the CCD images (as explained inthe main web page).

This figure shows the RMS residual of the blazar's brightness with respect to each night's model fit of the blazar's brightness to a model for extinction and image bias.  In the main web page it is argued that some of this variation within a night (~10 mMag) is due to actual blazar brightness variations on hourly time scales.

Return to W Com main web page

This site opened:  June 16, 2003.  Last Update:  June 22, 2003