From 3b24128b3cf7f3e648065f7570d94df1e4b914df Mon Sep 17 00:00:00 2001 From: pciturri Date: Tue, 26 Nov 2024 19:27:29 +0100 Subject: [PATCH] docs: Added resampled M-test and MLL test to the concepts section. Left Serafini et al., 2024 as in-prep, until pre-print is submitted. --- docs/concepts/evaluations.rst | 7 ++++++- docs/getting_started/theory.rst | 4 ++-- 2 files changed, 8 insertions(+), 3 deletions(-) diff --git a/docs/concepts/evaluations.rst b/docs/concepts/evaluations.rst index 05e94ad3..4e26ae81 100644 --- a/docs/concepts/evaluations.rst +++ b/docs/concepts/evaluations.rst @@ -30,7 +30,7 @@ consistency tests and they verify whether a forecast in consistent with an obser that can be used to compare the performance of two (or more) competing forecasts. PyCSEP implements the following evaluation routines for grid-based forecasts. These functions are intended to work with :class:`GriddedForecasts` and :class:`CSEPCatalogs``. -Visit the :ref:`catalogs reference` and the :ref:`forecasts reference` to learn +Visit the :ref:`catalogs reference` and the :ref:`forecasts reference` to learn more about to import your forecasts and catalogs into PyCSEP. .. note:: @@ -105,6 +105,8 @@ Consistency tests magnitude_test pseudolikelihood_test calibration_test + resampled_magnitude_test + MLL_magnitude_test Publication reference ===================== @@ -114,6 +116,8 @@ Publication reference 3. Magnitude test (:ref:`Savran et al., 2020`) 4. Pseudolikelihood test (:ref:`Savran et al., 2020`) 5. Calibration test (:ref:`Savran et al., 2020`) +6. Resampled Magnitude Test (Serafini et al., in-prep) +7. MLL Magnitude Test (Serafini et al., in-prep) **************************** Preparing evaluation catalog @@ -121,6 +125,7 @@ Preparing evaluation catalog The evaluations in PyCSEP do not implicitly filter the observed catalogs or modify the forecast data when called. For most cases, the observation catalog should be filtered according to: + 1. Magnitude range of the forecast 2. Spatial region of the forecast 3. Start and end-time of the forecast diff --git a/docs/getting_started/theory.rst b/docs/getting_started/theory.rst index 0b9362a4..24252a2c 100644 --- a/docs/getting_started/theory.rst +++ b/docs/getting_started/theory.rst @@ -1019,7 +1019,7 @@ on the figure by default. Resampled Magnitude Test ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -Aim: Perform the resampled magnitude test for catalog-based forecasts (Serafini et al., 2024), +Aim: Perform the resampled magnitude test for catalog-based forecasts (Serafini et al., , in-prep), which is a correction to the original M-test implementation that is biased to the total N of events. @@ -1064,7 +1064,7 @@ horizontal line. The quantile score for this forecast is Modified Multinomial Log-Likelihood (MLL) Magnitude Test ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -* Aim: Implements the modified Multinomial log-likelihood (MLL) magnitude test (Serafini et al., 2024). +* Aim: Implements the modified Multinomial log-likelihood (MLL) magnitude test (Serafini et al., in-prep). * Method: Calculates the test statistic distribution as: