Under the assumption that the trace statistic is equal to 6.7905, the critical value is equal to 15.4947 at the 5% significance level.
A minimum-weight, r-regular subgraph is constructed, and the mean cross-count test statistic is equal to the number of edges in the subgraph containing one observation from the first group and one from the second, divided by r.
Finally, the significance of all correlations was determined using Student's t-test, where the test statistic is equal to <img src="http://journals.plos.org/plosone/article/asset?id=info?doi/10.1371/journal.pone.0001296.e003.PNG" class= inline-graphic"/>, where r is the correlation coefficient and N is the number of samples (in this case, 19 countries).
For perfectly uncorrelated residuals the test statistic is equal to two.
Specifically, our RS method's test statistic is equal to the negative of the sum of the log p-values for each gene assigned to the pathway [ 10].
The counts of positive and negative signs in the acute renal failure in sepsis example were N+ = 13 and N- = 3, and S (the test statistic) is equal to the smaller of these (i.e. N-).
In particular, if one fits a parabola to the log likelihood at the null hypothesis (a good approximation locally), then the resulting LR test statistic is equal to a score statistic, for a score statistic that uses the observed rather than the more traditional expected information (Buse, 2007).
The stability statistic for a signature size k is defined as S t a b (k ) = ∑ i = 1 k f r e q (p i ) k m The Stab statistic is equal to 1 if the same signature is always selected over M10FOLDCV given a signature size and 1 m if there is no overlap.
The ROC AUC was equal to 0.90 and the H-L statistic was equal to 0.003 (p > 0.40).
The Durbin-Watson statistic was equal to 2.00, a value within the acceptable range of values from 1.50 to 2.50.
