Crandore Hub

modelimportance

Measuring Contributions of Component Models to Ensemble Forecast Accuracy

Provides metrics for quantifying the contribution of individual component models to the predictive accuracy of ensemble forecasts. The package implements the Leave-One-Model-Out (LOMO) and Leave-All-Subset-of-One-Model-Out (LASOMO) model importance metrics, enabling users to assess the relative importance of component models and better understand the performance of ensemble forecasting systems. Methods are described in Kim et al. (2026) <doi:10.1016/j.ijforecast.2025.12.006>.

README

# inst Directory

## `get-testdata` 

This directory contains scripts to generate test datasets used in the package's 
unit tests. The test datasets are stored in the sub-directories under 
`tests/testthat/testdata`, with each subdirectory named after the function being
tested.

* `complex_hub.R` generates test data for the `validate_input_data()` and 
`split_data_by_task()` functions.
* Test data sets for the `compute_importance()` function are generated from:
  * four top-level scripts matching the pattern `exp_imp_*_lomo.R`
  * scripts under `exp_imp_*_lasomo/` directories
  * a helper script, `helper-exp_imp.R`
* Test data for the `model_importance()` function is generated from: 
  * `forecast-target-datasets.R`
  * `exp_all_imp_scores.R`
  * a helper script, `helper-exp_all_imp_scores.R`

Versions across snapshots

VersionRepositoryFileSize
0.1.0 rolling linux/jammy R-4.5 modelimportance_0.1.0.tar.gz 2.0 MiB
0.1.0 rolling linux/noble R-4.5 modelimportance_0.1.0.tar.gz 2.0 MiB
0.1.0 rolling source/ R- modelimportance_0.1.0.tar.gz 3.8 MiB
0.1.0 latest linux/jammy R-4.5 modelimportance_0.1.0.tar.gz 2.0 MiB
0.1.0 latest linux/noble R-4.5 modelimportance_0.1.0.tar.gz 2.0 MiB
0.1.0 latest source/ R- modelimportance_0.1.0.tar.gz 3.8 MiB
0.1.0 2026-04-23 source/ R- modelimportance_0.1.0.tar.gz 0 B

Dependencies (latest)

Imports

Suggests