SETAR-Tree: a novel and accurate tree algorithm for global time series forecasting

DOI10.1007/S10994-023-06316-XarXiv2211.08661OpenAlexW4324026557MaRDI QIDQ6134328FDOQ6134328

Authors: Rakshitha Godahewa, Geoffrey I. Webb, Daniel R. Schmidt, Christoph Bergmeir 

Publication date: 22 August 2023

Published in: Machine Learning (Search for Journal in Brave)

Abstract: Threshold Autoregressive (TAR) models have been widely used by statisticians for non-linear time series forecasting during the past few decades, due to their simplicity and mathematical properties. On the other hand, in the forecasting community, general-purpose tree-based regression algorithms (forests, gradient-boosting) have become popular recently due to their ease of use and accuracy. In this paper, we explore the close connections between TAR models and regression trees. These enable us to use the rich methodology from the literature on TAR models to define a hierarchical TAR model as a regression tree that trains globally across series, which we call SETAR-Tree. In contrast to the general-purpose tree-based models that do not primarily focus on forecasting, and calculate averages at the leaf nodes, we introduce a new forecasting-specific tree algorithm that trains global Pooled Regression (PR) models in the leaves allowing the models to learn cross-series information and also uses some time-series-specific splitting and stopping procedures. The depth of the tree is controlled by conducting a statistical linearity test commonly employed in TAR models, as well as measuring the error reduction percentage at each node split. Thus, the proposed tree model requires minimal external hyperparameter tuning and provides competitive results under its default configuration. We also use this tree algorithm to develop a forest where the forecasts provided by a collection of diverse SETAR-Trees are combined during the forecasting process. In our evaluation on eight publicly available datasets, the proposed tree and forest models are able to achieve significantly higher accuracy than a set of state-of-the-art tree-based algorithms and forecasting benchmarks across four evaluation metrics.

Full work available at URL: https://arxiv.org/abs/2211.08661

Recommendations

• TS-CHIEF: a scalable and accurate forest algorithm for time series classification
• Dynamic staged trees for discrete multivariate time series: forecasting, model selection and causal analysis
• A time series forest for classification and feature extraction
• scientific article; zbMATH DE number 1750491
• Forecasting methods for time series with large alphabets based on universal measure and decision trees
• Modeling of time series using random forests: theoretical developments
• Interval forecasts based on regression trees for streaming data
• A novel hybrid ARIMA and regression tree model for the interval-valued time series

zbMATH Keywords

threshold autoregressive modelstree modelsforest modelsglobal time series forecasting

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Cites Work

• Statistical Analysis of Financial Data in S-Plus
• Greedy function approximation: A gradient boosting machine.
• Generalized random forests
• Random forests
• Specification, estimation, and evaluation of smooth transition autoregressive models
• Title not available (Why is that?)
• The Lagrange Multiplier Test and its Applications to Model Specification in Econometrics
• Title not available (Why is that?)
• Forecasting with exponential smoothing. The state space approach
• Oscillation and Chaos in Physiological Control Systems
• Simple mathematical models with very complicated dynamics
• Title not available (Why is that?)
• Applied Predictive Modeling
• Smooth transition autoregressive models and fuzzy rule-based systems: Functional equivalence and consequences
• NON-NORMALITY AND TESTS ON VARIANCES
• Exchange rate forecasting: Results from a threshold autoregressive model
• Linearity testing for fuzzy rule-based models
• The behaviour of US stock prices: Evidence from a threshold autoregressive model
• Tree-structured smooth transition regression models
• A Threshold Autoregressive Model for Wholesale Electricity Prices
• Exploring the sources of uncertainty: why does bagging for time series forecasting work?
• An explicit split point procedure in model-based trees allowing for a quick fitting of GLM trees and GLM forests
• Arbitrage of forecasting experts
• Self Exciting Threshold Autoregressive Models for Describing Cyclical Data

This page was built for publication: SETAR-Tree: a novel and accurate tree algorithm for global time series forecasting

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6134328)