csep.core.poisson_evaluations.conditional_likelihood_test

csep.core.poisson_evaluations.conditional_likelihood_test(gridded_forecast, observed_catalog, num_simulations=1000, seed=None, random_numbers=None, verbose=False)[source]

Performs the conditional likelihood test on Gridded Forecast using an Observed Catalog.

This test normalizes the forecast so the forecasted rate are consistent with the observations. This modification eliminates the strong impact differences in the number distribution have on the forecasted rates.

Note: The forecast and the observations should be scaled to the same time period before calling this function. This increases transparency as no assumptions are being made about the length of the forecasts. This is particularly important for gridded forecasts that supply their forecasts as rates.

Parameters:
  • gridded_forecast – csep.core.forecasts.GriddedForecast

  • observed_catalog – csep.core.catalogs.Catalog

  • num_simulations (int) – number of simulations used to compute the quantile score

  • seed (int) – used fore reproducibility, and testing

  • random_numbers (numpy.ndarray) – random numbers used to override the random number generation. injection point for testing.

Returns:

csep.core.evaluations.EvaluationResult

Return type:

evaluation_result