A 'modeltime' extension that implements time series ensemble forecasting methods including model averaging,
weighted averaging, and stacking. These techniques are popular methods
to improve forecast accuracy and stability. Refer to papers such as
"Machine-Learning Models for Sales Time Series Forecasting" Pavlyshenko, B.M. (2019) <doi:10.3390>.
| Version: |
1.0.1 |
| Depends: |
modeltime (≥ 1.2.2), modeltime.resample (≥ 0.2.1), R (≥
3.5) |
| Imports: |
tune (≥ 0.1.2), rsample, yardstick, workflows (≥ 0.2.1), parsnip (≥ 0.1.6), recipes (≥ 0.1.15), timetk (≥ 2.5.0), tibble, dplyr (≥ 1.0.0), tidyr, purrr, glue, stringr, rlang (≥ 0.1.2), cli, generics, magrittr, tictoc, parallel, doParallel, foreach |
| Suggests: |
gt, crayon, dials, glmnet, progressr, utils, roxygen2, earth, testthat, tidymodels, xgboost, tidyverse, lubridate, knitr, rmarkdown, covr, qpdf, remotes |
| Published: |
2022-06-09 |
| Author: |
Matt Dancho [aut, cre],
Business Science [cph] |
| Maintainer: |
Matt Dancho <mdancho at business-science.io> |
| BugReports: |
https://github.com/business-science/modeltime.ensemble/issues |
| License: |
MIT + file LICENSE |
| URL: |
https://github.com/business-science/modeltime.ensemble |
| NeedsCompilation: |
no |
| Materials: |
README NEWS |
| In views: |
TimeSeries |
| CRAN checks: |
modeltime.ensemble results |