Formal Results: Testing the GCP Hypothesis

A Bottom Line Graph

pictures the history of accumulated deviations from expectation.

The Hypothesis Registry

defines formal events. Event analyses are linked in Table 2 below.

Exploratory Studies

provide additional insight, and help develop formal hypotheses.

Introduction

More than 15 years after its inception, the GCP continues to gather data. There are over 60 active eggs in the network, in Europe, the US, Canada, India, Fiji, New Zealand, Japan, China, Russia, Brazil, Africa, Thailand, South America, Australia. We hypothesize that our instrument (the network of eggs) will show anomalous deviations associated with Global Events when there is widespread participation or reaction to the event.

Specific tests of the hypothesis are formalized using a regular set of required specifications. The test events are entered into a table in our Formal Hypothesis Registry which provides a priori identification and complete specification of all parameters we use for statistical analysis and inferences.

Table 1 shows the composite result for all formally specified events.

Table 2 shows the results for individual events. Table entries for most events have direct links to a description and analysis. The events are identified by name and date, and a column with the heading, Hypothesis Source, indicates who made the prediction. The number of contributing REGs is shown, and the Resolution, which indicates the block-size for the analysis. The last two columns show the Z-score statistic for the analysis, and its associated probability. A few early events were poorly defined or did not use standard analysis–they are marked with red asterisks (*). Statistics for very recent events (marked with ∼) will change slightly when normalized calculations are done. The table is sortable by columns.

Following Table 2, the composite numerical results are presented visually. The evidence for the general hypothesis is shown in a scatterplot and in a chronological graph of the accumulating formal data. In the latter display, the terminal value corresponds to the overall Z-score.

Table 1: Overall Summary of Results
August 1998 to December 2015

Table 2: Results for Individual Events

red

: significant

light red

: predicted direction

green

: opposite & significant

Graph of Accumulating Deviations: The Complete Formal Database

The two following figures represent the history of our formal hypothesis testing. The first shows the Z-scores for almost 500 formally specified events in an ordinary scatterplot. While there is a noticeable positive bias, it is not easy to see its significance. Yet the odds against chance of this meanshift over a database this size are about a hundred billion to one.

The second figure displays the same data as a cumulative deviation from chance expectation (shown as the horizontal black line at 0 deviation). Truly random data would produce a jagged curve with no slope, wandering up and down around the horizontal. The dotted smooth curves show the 0.05 and 0.001 and 0.000001 probability envelopes that indicate significant versus chance excursions. This figure can be compared with a control distribution using simulations of the event series.

The jagged red line shows the accumulating excess of the empirically normalized Z-scores relative to expectation for the complete dataset of rigorously defined events. The overall result is highly significant. The odds against chance are more than a trillion to one.