[Aavso-photometry] Transforms: what about VERR?

[arne]

George Silvis wrote:
> When you transform your VMAG information, what should you do with 
> VERR? Is it unchanged by the transform? Or should it reflect the 
> uncertainty of the transform coefficients? Suggestions?
> 
> I have put together an application to apply transform coefficients ( 
> it is available to all at http://www.gasilvis.net/transformer/ ). 
> Right now it does not modify the VERR, but it would be good to do the 
> right thing.
> 
There are two types of errors: systematic, such as the transformation
errors; and random, such as for the actual photometric measures.
If I am just reporting photometry of variable stars, I just report
the random error, and include as a note that an additional error
of x.xxx needs to be added in quadrature for the error in
transforming to the standard system.  If I am calibrating a field,
the reported errors include the transformation determination error
since now the accuracy of the standard magnitude is important.

In theory, your errors should include *all* known error terms.
It is sobering when you start including everything you can think of.
However, for time-series analysis, including systematic errors that
affect every data point about equally overstates the error as far
as differential photometry is concerned.
Arne