Probabilistic forecasting with temporal convolutional neural network

Yitian Chen, Yanfei Kang, Yixiong Chen, Zizhuo Wang

Research output: Contribution to journalArticlepeer-review

244 Scopus citations

Abstract

We present a probabilistic forecasting framework based on convolutional neural network (CNN) for multiple related time series forecasting. The framework can be applied to estimate probability density under both parametric and non-parametric settings. More specifically, stacked residual blocks based on dilated causal convolutional nets are constructed to capture the temporal dependencies of the series. Combined with representation learning, our approach is able to learn complex patterns such as seasonality, holiday effects within and across series, and to leverage those patterns for more accurate forecasts, especially when historical data is sparse or unavailable. Extensive empirical studies are performed on several real-world datasets, including datasets from JD.com, China's largest online retailer. The results show that our framework compares favorably to the state-of-the-art in both point and probabilistic forecasting.

Original languageEnglish (US)
Pages (from-to)491-501
Number of pages11
JournalNeurocomputing
Volume399
DOIs
StatePublished - Jul 25 2020

Bibliographical note

Funding Information:
Yanfei Kang’s research were supported by the National Natural Science Foundation of China (No. 11701022 ).

Publisher Copyright:
© 2020 Elsevier B.V.

Keywords

  • Convolutional neural network
  • Demand forecasting
  • Dilated causal convolution
  • High-dimensional time series
  • Probabilistic forecasting

Fingerprint

Dive into the research topics of 'Probabilistic forecasting with temporal convolutional neural network'. Together they form a unique fingerprint.

Cite this