Hi Based on what I've seen in terms of daily / weekly / monthly correlation in data - a yearly correlation would not surprise me at all. Coming up with a test environment that's immune to external influence is not at all easy. The NIST guys worry quite a bit about stuff that's sitting in caves. Periodicity in data is very common. Digging down and finding the reason for that periodicity - not so common.
Bob On Nov 14, 2010, at 11:13 AM, Magnus Danielson wrote: > On 11/14/2010 04:05 PM, [email protected] wrote: >> Hi all, >> >> thanks for your comments. >> >> I take from them that the "galactic" jitter in an oscillator >> can't be seen unless one has a long time series (such as I did >> with temperature). Too many other causes would mask it, as >> some of you have evidenced. Nevertheless it exists, but has >> no practical implications in the current practice at our labs. >> >> Should anybody have an interest in my curve (maybe using it as >> a reference....), it is at >> >> http://xoomer.virgilio.it/iovane/trimestri1.xls >> >> Please look mainly at the curve labeled "ALL (A to H)", which >> summarizes two years of data (6+ million data points). Notice >> the valley when I'm opposite to the center of the galaxy. > > What are the scales? > What are the time-reference? > > If you have shown that the feature has a 86164 second period rather than > 86400 s period a good exercise would be to show that it has a high > correlation to the integral of the half-hemisphere gamma rays (as show in the > graphs) the experiment is facing. Some deviation may naturally be expected, > as the experiment may not have the same sensitivity in all directions to > gamma rays, but the basic correlation should be there. > This correlation could be made into a stronger proof if done over the year, > as the hemisphere shifts over the sky over the years due to the angle of the > earth. > > Anyway, I think we are going into off-topicness here. > > Cheers, > Magnus > > _______________________________________________ > time-nuts mailing list -- [email protected] > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
