csep.core.poisson_evaluations.w_test

csep.core.poisson_evaluations.w_test(gridded_forecast1, gridded_forecast2, observed_catalog, scale=False)[source]

Calculate the Single Sample Wilcoxon signed-rank test between two gridded forecasts.

This test allows to test the null hypothesis that the median of Sample (X1(i)-X2(i)) is equal to a (N1-N2) / N_obs. where, N1, N2 = Sum of expected values of Forecast_1 and Forecast_2, respectively.

The Wilcoxon signed-rank test tests the null hypothesis that difference of Xi and Yi come from the same distribution. In particular, it tests whether the distribution of the differences is symmetric around given mean.

Parameters

Parameters:

gridded_forecast1 – Forecast of a model_1 (Grided) (Numpy Array) A forecast has to be in terms of Average Number of Events in Each Bin It can be anything greater than zero
gridded_forecast2 – Forecast of model_2 (Grided) (Numpy Array) A forecast has to be in terms of Average Number of Events in Each Bin It can be anything greater than zero
observation – Observed (Grided) seismicity (Numpy Array): An Observation has to be observed seismicity in each Bin It has to be a either zero or positive integer only (No Floating Point)

Returns: out: csep.core.evaluations.EvaluationResult