R2N2: Residual Recurrent Neural Networks for Multivariate Time Series Forecasting
This addresses the problem of efficient and accurate forecasting for applications like aviation and climate, but it is incremental as it builds on existing VAR and RNN methods.
The paper tackled multivariate time series forecasting by proposing R2N2, a hybrid model combining linear VAR and nonlinear RNN components to improve predictive performance and reduce training complexity, showing it is competitive and often better than using VAR or RNN alone on aviation and climate datasets.
Multivariate time-series modeling and forecasting is an important problem with numerous applications. Traditional approaches such as VAR (vector auto-regressive) models and more recent approaches such as RNNs (recurrent neural networks) are indispensable tools in modeling time-series data. In many multivariate time series modeling problems, there is usually a significant linear dependency component, for which VARs are suitable, and a nonlinear component, for which RNNs are suitable. Modeling such times series with only VAR or only RNNs can lead to poor predictive performance or complex models with large training times. In this work, we propose a hybrid model called R2N2 (Residual RNN), which first models the time series with a simple linear model (like VAR) and then models its residual errors using RNNs. R2N2s can be trained using existing algorithms for VARs and RNNs. Through an extensive empirical evaluation on two real world datasets (aviation and climate domains), we show that R2N2 is competitive, usually better than VAR or RNN, used alone. We also show that R2N2 is faster to train as compared to an RNN, while requiring less number of hidden units.