Recurrent Neural Networks for Time Series Forecasting: Current Status and Future Directions
By Hansika Hewamalage, Christoph Bergmeir, Kasun Bandara. International Journal of Forecasting, 2019
1,252 citations recorded by the source corpus.
Abstract Recurrent Neural Networks (RNNs) have become competitive forecasting methods, as most notably shown in the winning method of the recent M4 competition. However, established statistical models such as exponential smoothing (ETS) and the autoregressive integrated moving average (ARIMA) gain their popularity not only from their high accuracy, but also because they are suitable for non-expert users in that they are robust, efficient, and automatic. In these areas, RNNs have still a long way to go. We present an extensive empirical study and an open-source software framework of existing RNN architectures for forecasting, and we develop guidelines and best practices for their use. For example, we conclude that RNNs are capable of modelling seasonality directly if the series in the dataset possess homogeneous seasonal patterns; otherwise, we recommend a deseasonalisation step. Comparisons against ETS and ARIMA demonstrate that (semi-) automatic RNN models are not silver bullets, but they are nevertheless competitive alternatives in many situations.
Subfields: Electrical and Electronic Engineering, Management Science and Operations Research
View Paper: https://www.semanticscholar.org/paper/3e707aa0d02b637336739e77f3b05e4fdebb4fba
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Electrical and Electronic Engineering
In Management Science and Operations Research
- Bernie Wang
- Tim Januschowski
- Syama Sundar Rangapuram
- Jan Gasthaus
- Laurent Callot
- Danielle C. Maddix
- David Salinas
- Michael Bohlke-Schneider
Citations in the corpus
- Deep Learning for Time Series Forecasting: Tutorial and Literature Survey (2020, methodology): “[76] provide an overview specifically targeted at RNNs for forecasting.”
- An Experimental Review on Deep Learning Architectures for Time Series Forecasting (2020, background): “Existing reviews have either focused just on a specific type of deep learning architecture 12 or on a particular data scenario.”
- A critical review of RNN and LSTM variants in hydrological time series predictions (2024, methodology): “Solutions to these limitations involve using regularization techniques like dropout to reduce overfitting, employing data augmentation to enhance training sets, and optimizing hyperparameters through cross-validation [ 120 ].”