The flare likelihood and region eruption forecasting (FLARECAST) project: flare forecasting in the big data & machine learning era | Journal of Space Weather and Space Climate

Topical Issue - Space Weather research in the Digital Age and across the full data lifecycle

Open Access

Table 4

Metrics and skill scores used for categorical forecasts in FLARECAST, along with their applicable ranges. All parameters shown rely on the simple 2 × 2 contingency table of Table 3.

Name	Notation	Formula	Range
Accuracy	ACC		[0, 1]
False alarm ratio	FAR		[0, 1]
Bias	BIAS		[0, ]
Threat score	TS		[0, 1]
Equitable threat score	ETS		[−, 1]
Equitable threat score	ETS	Using
Probability of detection	POD		[0, 1]
Probability of false detection	POFD		[0, 1]
Odds ratio	OR		[0, ]
Odds ratio skill score	ORSS		[−1, 1]
Heidke skill score	HSS		[−1, 1]
Heidke skill score	HSS	Using
True skill statistic	TSS	POD − POFD	[−1, 1]
Symmetric extremal dependence index	SEDI		[−1, 1]
Appleman’s discriminant	AD	if (TP + FN) > (FP + TN)	[−, 1]
Appleman’s discriminant	AD	if (TP + FN) < (FP + TN)	[−, 1]

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.