New Year Devvar-Netvar Analysis

In 2005, a "data-mining" assessment of seven years of data around the New Year transition was undertaken to develop a new hypothesis to test on future data. Two recipes were devised, which differ in detail but are parallel to the original analyses done in previous years. This page gives results for the equivalent to the "standard analysis" which was an epoch average of the cumulative deviation of Chisquare in each timezone.

The 2006 recipe is an epoch average over 13 timezones of the difference of two cumulative deviation curves. One is the the Devvar (device variance), which our retrospective analysis showed tends to have a positive deviation. From this we subtract the Netvar cumulative (the Stouffer Z², which tends to have a negative deviation. Thus, our prediction is a positive trend for Devvar, and a negative trend for Netvar, and the test is a simple algebraic difference, Devvar-Netvar.

We also determined that only a few timezones contribute to the deviations, and not surprisingly, these are the heavily populated zones, and especially those that tend to actually celebrate the conventional Western New Year at midnight on December 31st. The datamining identified 13 timezones, namely {-8, -7, -6, -5, -4, -3, 0, 1, 2,3, 5.5,8, 9}. This includes the US and European zones and a few others.

For 2006, the result for this analysis is not positive, with a Z-score of -0.510, and p=0.695. Though this is substantially different from the average over the past several years, the difference is not significant. Despite the lack of confirmation, we will continue to use this new prediction in coming years. In principle, the likelihood of this test showing positive results should be high.

I will provide a graphic version of the analysis when available.


GCP Home