posted on Aug, 21 2013 @ 01:33 AM
reply to post by BayesLike
Here's a little chart with the low count days highlighted in red. The vertical axis is in Ln(counts), which works better than just the counts.
SQRT(counts) may or may not be more appropriate for getting a symmetric chart. I'll have to look into that with a longer dataset. You can see that
the big count days do not appear follow these low count days. But there aren't very many of them.... I suspect the big count days occur on or just
after major EQs.
There's maybe a hint that these low count days are the beginning of a typically multimonth period of increasing counts at least once or twice. It's
not at all clear with this tiny dataset over a short time period that this will hold up either. I might try pulling a few decades of data if the
server will let me do that. But that would be another posting.
Darn, the server balks at more than 20k records, and I need about 550k records. I'll try to find out if there is another way to get the longer
dataset.
edit on 21-8-2013 by BayesLike because: (no reason given)
edit on 21-8-2013 by BayesLike because: (no reason
given)
edit on 21-8-2013 by BayesLike because: (no reason given)