Use of the likelihood for measuring the skill of probabilistic forecasts

Abstract

We define the likelihood and give a number of justifications for its use as a skill measure for probabilistic forecasts. We describe a number of different scores based on the likelihood, and briefly investigate the relationships between the likelihood, the mean square error and the ignorance.

Version 1 (3rd August 2003) contained some incorrect statements about the relationship between the likelihood and the Brier score. These have now been removed

References in corpus (1)

• The Assessment and Calibration of Ensemble Seasonal Forecasts of Equatorial Pacific Ocean Temperature and the Predictability of Uncertainty

Cited by in corpus (6)

• Moment based methods for ensemble assessment and calibration
• The problem with the Brier score
• Improving probabilistic weather forecasts using seasonally varying calibration parameters
• Probabilistic forecasts of temperature: measuring the utility of the ensemble spread
• Probabilistic temperature forecasting: a comparison of four spread-regression models
• Five guidelines for the evaluation of site-specific medium range probabilistic temperature forecasts