Steve Jobs Dies
Steve Jobs was 56. He died on October 5 2011 after publicly fighting advanced pancreatic cancer since 2004. He, along with co-founder Steve Wozniak, built one of the most commercially successful personal computers and founded Pixar animation studios.
He was a creative visionary who changed the world of technology by inspiring and building elegant, highly functional, literally beautiful machines. The Mac computer, the PowerBook, the iPod, iPhone, iPad. Legions of ordinary people were given access to powerful creativity and communication devices that worked -- simply and reliably -- while also looking good.
The Apple Board of Directors wrote:
You can share remembrances with Apple at firstname.lastname@example.org. Here is a post from the eve of his stepping down as CEO.
Many people wondered if the GCP network might respond to the passing of Steve Jobs. I was not able to find a time of death, but it probably was in the morning of the 5th, local time. The GCP event was set for 10:00 to 16:00 local time (17:00-23:00 UTC). The result is Chisquare 22048.268 on 21600 df for p = 0.016 and Z = 2.145.
An exploratory look at the context for this formal event is interesting. As noted, we do not have the time of death, and also had little concrete information about the timing of announcements, statements, press releases. The Internet was quickly flooded with information and commentary beginning sometime in the afternoon. The following figure shows the full 24-hour UTC day, with the period of the Formal Event marked. It is an unusual dataset, with two periods of consistent deviation from expectation, one negative and one positive, creating the two long trends in the cumulative deviation graph. As always, we note that no reliable interpretations can be made of single events or individual analyses.
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.