Evaluating probabilistic forecasts with scoringRules

TL;DR

The paper introduces the scoringRules package for R, which facilitates the evaluation and comparison of probabilistic forecasts across various fields using proper scoring rules.

Contribution

It provides a comprehensive implementation of proper scoring rules in R, along with practical case studies demonstrating its application in meteorology and economics.

Findings

Effective evaluation of probabilistic models using scoringRules

Case studies illustrate practical usage in meteorology and economics

Supports diverse probabilistic forecast assessments

Abstract

Probabilistic forecasts in the form of probability distributions over future events have become popular in several fields including meteorology, hydrology, economics, and demography. In typical applications, many alternative statistical models and data sources can be used to produce probabilistic forecasts. Hence, evaluating and selecting among competing methods is an important task. The scoringRules package for R provides functionality for comparative evaluation of probabilistic models based on proper scoring rules, covering a wide range of situations in applied work. This paper discusses implementation and usage details, presents case studies from meteorology and economics, and points to the relevant background literature.