The time series forecasting package for the tidymodels
ecosystem.
No need to switch back and forth between various frameworks. modeltime
unlocks machine learning & classical time series analysis.
- forecast: Use ARIMA, ETS, and more models coming (
arima_reg()
,arima_boost()
, &exp_smoothing()
). - prophet: Use Facebook’s Prophet algorithm (
prophet_reg()
&prophet_boost()
) - tidymodels: Use any
parsnip
model:rand_forest()
,boost_tree()
,linear_reg()
,mars()
,svm_rbf()
to forecast
Modeltime incorporates a simple workflow (see Getting Started with Modeltime) for using best practices to forecast.
All plots incorporate both plotly
(interactive) and ggplot2
(static)
visualizations. This means you can quickly add forecasts to shiny
apps, rmarkdown
documents, and more.
-
Getting Started with Modeltime: A walkthrough of the 6-Step Process for using
modeltime
to forecast -
Modeltime Documentation: Learn how to use
modeltime
, find Modeltime Models, and extendmodeltime
so you can use new algorithms inside the Modeltime Workflow.
Install the development version from with:
# install.packages("devtools")
devtools::install_github("business-science/modeltime")
I teach modeltime
in my Time Series Analysis & Forecasting Course.
If interested in learning Pro-Forecasting Strategies then join my
waitlist.
The course is coming soon.
You will learn:
- Time Series Preprocessing, Noise Reduction, & Anomaly Detection
- Feature engineering using lagged variables & external regressors
- Hyperparameter Tuning
- Time series cross-validation
- Ensembling Multiple Machine Learning & Univariate Modeling Techniques (Competition Winner)
- NEW - Deep Learning with RNNs (Competition Winner)
- and more.