Libyan Turmoil Continues
The ongoing battles in Libya continue, with no resolution in sight. It is a situation of interest around the world, and the attacks and counterattacks on the 2nd of March seemed to be an appropriate sample of the turmoil. We therefore set a formal event that might cover most of the major battles on that day. (Notes extracted from Globe and Mail reporting.)
The 2nd was also marked by an apparent humanitarian gesture by Gadhafi forces, who delivered food and water to the refugees blocked at the Tunisian border. They also had reclaimed the eastern cities of Brega and Ajdabiya, pushing ever closer to the rebel capital of Benghazi.
However, by the end of the day, fortunes in the east had turned again as the revolutionaries and pro-Gadhafi forces fought a day-long battle with heavy artillery and assault rifles, ending with the opposition forces -- a mix of defecting Libyan officers and armed activists -- claiming control of the town. But government air strikes continued into the night, and there were widespread reports of heavily armed government troops on the move.
The most serious loss for the opposition occurred out of sight of the world's media in the highly populated west of Libya. There, Gadhafi forces moved the highly armed Khweldi tank and infantry brigade to the town of Sorman and placed an even larger brigade to the immediate west of Tripoli, surrounding the opposition-held city of Zawiyah, according to independent monitors and Libyans living in the cities.
I was not able to find any times associated with the major clashes. The GCP event was set for 12 noon to midnight local Tripoli time on the 2nd of March. The result is 43638.506 on 43200 df, for p = 0.068 and Z = 1.489.
An informal exploration looking at the full UTC day (which covers 2 AM Wednesday to 2 AM Thursday local time) shows pretty much the same picture, with a number of periods of strong, persistent deviation, interspersed with relatively flat periods.
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.