Re: Tass@Home

[Tom Droege <tdroege2@earthlink.net>]

Doug and all,

To give you an idea of the magnitude of this problem, I now have 375 disks 
in the stack.  The data consists of multiple 56 measurement sequences 
spaced by one to 100 days.   I am now keeping fairly accurate time.  Most 
images are timed to within 2 seconds for the 100 second +/- 1/2 second 
exposure.  The 56 measurements are taken over 2.3 hours.

The time series analysis has to cope with groups of closely spaced 
measurements spaced by days.

You bet, I think we should analyze every time sequence.  This is obviously 
something that could be sent out over the internet.  Also the code is 
probably small, and well understood??  Doug, lets assume a data set 
consisting of 5 days of observations spaced out over several months, with 
56 observations (in V and I) per day.  How long would this grind on a 1 GHz 
computer?  What do bad points do to the result?  i.e. will a real period of 
small amplitude win out over a false period caused by a couple of bad 
points.  In the frequency domain a couple of false points might show up as 
a sharp spike on a more rounded background??

There might be several hundred thousand time sequences in the engineering 
data set that I have.

Grinding just the data that I already have through a program to get the 
star lists is a big job and might be divided up among an analysis group.

Tom Droege


At 07:27 PM 1/29/02 -0500, you wrote:

>Tom et al,
>
>I think that one can very probably keep up with the photometry
>at a single site with a single modern computer. The task
>of time-series analysis on *every* available star would be
>unprecedented and would likely lead to unanticipated findings.
>I am therefore more in favor of farming out lightcurves, perhaps
>in groups of dozens, for distributed time-series analysis.
>
>Cheers,
>Doug