setartree

The implementation of a forecasting-specific tree-based model that is in particular suitable for global time series forecasting, as proposed in Godahewa et al. (2022) doi:10.48550/arXiv.2211.08661. The model uses the concept of Self Exciting Threshold Autoregressive (SETAR) models to define the node splits and thus, the model is named SETAR-Tree. The SETAR-Tree uses some time-series-specific splitting and stopping procedures. It trains global pooled regression models in the leaves allowing the models to learn cross-series information. The depth of the tree is controlled by conducting a statistical linearity test as well as measuring the error reduction percentage at each node split. Thus, the SETAR-Tree requires minimal external hyperparameter tuning and provides competitive results under its default configuration. A forest is developed by extending the SETAR-Tree. The SETAR-Forest combines the forecasts provided by a collection of diverse SETAR-Trees during the forecasting process.

Last 30 days

Last 365 days

The following line graph shows the downloads per day. You can hover over the graph to see the exact number of downloads per day.