Re: [Phys-L] Coincidence Statistics
> I wrote a quick python program to do a little Monte Carlo. It always gets
Can you post the Python code or send it to me by email?
When doing coincidence work (e.g., Geiger events) I find it convenient
to work from raw timestamp data instead of counts or bins. If you
timestamp all events from source/channel A and timestamp all events from
source/channel B it's easy to then retroactively go through the data and
evaluate it with different heuristics.
In the precise timing world there isn't literal coincidence since you
can measure down to microseconds or nanoseconds and at that level all
timestamps tend to be unique. So the question then is what loose
definition of coincidence is to be used. If you use, say, 1 ms bins,
then you can have cases where two events are 990 us apart but still be
the same bin, or cases where two events are merely 20 us apart yet in
different bins. Or maybe in the whole scheme of things that's too minor
an issue to worry about? I don't really know. But I tend not to use
fixed bins and use time difference of arrival instead.
/tvb