Variance Deviations on September 11

The most striking departures of the GCP data from expectation on September 11 were in the variance measure, that is, the second-by-second internode variation. Beginning at about 4 or 5 in the morning, New York time, the variance showed a persistent, slight increase, that continued until about 11:00. It then became slightly smaller than expectation and again, maintained the tendency for several hours, until about 18:00. The following figures show the cumulative deviation of the variance from its mean value for one day, the 11th, and then the same measure in a larger context of three days and 12 days.

The first figure represents the formal, scientific view. Here we see very steep and long-lasting trends in the actual data, and nothing similar in the pseudo-random control data. This is the formal result, and the extreme excursion is unlikely with odds less than 1 in 1000.

Terrorist Attacks,
September 11 2001

The second and third figures, although they show the actual data, emphasize an aesthetic perspective, and give a thought provoking, broader context. Just as a picture, in a purely aesthetic interpretation, the next graph, shows Sept 11 in the context of three days. It looks like a long, emphatic moment of feeling and knowing by a global consciousness. This piercing graphic cry seems symbolic of the pain surrounding our slow and spasmodic struggle to become human. It is possible to envision a global consciousness that we may one day embrace, and that could embrace us all.

Variance, 9/11, 3

The extraordinary character of the deviations is even more obvious in the next figure, which plots the square of the data values. It shows the relative magnitude of the variance excursions on 9/11 against the background of surrounding days. (Larger fullscreen version) Cumdev Variance Sept 11

We can also look at a larger context. The impressionistic sense is that the September 11 spike stands on a general rise suggesting, in a creative interpretation, a larger scale of perception and a longer premonition. If the general picture of the graph is liberally taken, it seems to show that there was a period of several days leading to and following the 11th during which the EGG data showed unusual behavior. Of course, this graph also shows that the spike on the 11th is unique mainly by its place in the flow of time.
Variance, 9/11, 12

Finally, if we separate the data into groups by location, we can see that contributions to the general shape of the variance curve come from several, but not all areas of the world. The differences shown here are merely suggestive, not significant, but it is notable that the eggs on the east coast of the US have the largest contribution. On the other hand, eggs in the middle and west part of the US contribute less than those far away, for example in Brazil, Africa, India, New Zealand and Australia.
Variance, by location

GCP Home