Once ClusterSeer has performed a Dat's analysis, it writes information on the analysis and results into the session log.
Datasets used
Number of time series
Number of cells per series
The test statistic A, is the number of cells containing more than the number of cases expected in the absence of clustering.
ClusterSeer then reports the results of the analysis for the single and several time series, including the expected value of A
(E(A)), its variance (Var (A)), z-score, and Upper-tail P-value for each row in the time series. The upper-tail P-value is the probability under the null hypothesis of observing an A as large or larger than the one given in A. It is based on a normal distribution of the data.
For several time series, ClusterSeer also provides a Chi-squared statistic, an overall P-value, and a list of time series it was unable to analyze.
ClusterSeer will plot the observed statistic (A) on its expectation (E(A)) for a single and/or several time series. This plot describes where observations would be plotted under the null hypothesis of cases occurring at random over the t time periods.