|
The 5th of August was not a great day for the US. We were
braced for a possible downgrade of the US Government bond
rating from its AAA level, and late on Friday (I haven't
been able to find the actual time) Standard and Poor
announced a downgrade to AA+, the first time in history this
happened to the US.
Then, early on the 6th we heard about a helicopter crash in
Afghanistan during the night (described as late Friday),
with 31 special forces killed, and 7 or 8
Afghans, and all indications were that the Taliban had
brought it down with a rocket grenade. The news further
indicated that most of the Americans were from the
Navy Seal Team 6 that had famously killed Osama Bin Laden.
I originally set this event to cover the approximate time of the
helicopter crash plus 6 hours for the news to spread, but
then recognized that there was a coincidence with the
news of the US credit rating downgrade and the immediate plunge of stock
market futures. I decided to name the event for both of
these somewhat shocking news items, and to provide some
description to give GCP readers a flavor of the complexity
we deal with in identifying events for analysis. In fact,
such overlaps of potential GCP events are common, and we have
to depend on precise timing and limited duration of defined events
to help disentangle them.
In this case, especially because we don't have precise times
for either event, the outcome is ambiguous. There is no
significant trend, and the test statistic is modestly
negative. The result is Chisquare 21483.105 on 21600 df for
p = 0.712 and Z = -0.560.
It is important to keep in mind that we have only a tiny
statistical effect, so that it is always hard to distinguish
signal from noise. This means that every "success" might be
largely driven by chance, and every "null" might include a real
signal overwhelmed by noise. In the long run, a real effect can
be identified only by patiently accumulating replications of
similar analyses.
|