Forecasting Across Time Series Databases using Recurrent Neural Networks on Groups of Similar Series: A Clustering Approach

TL;DR

This paper introduces a clustering-based approach using RNNs, specifically LSTM networks, to improve forecasting accuracy across heterogeneous time series databases by grouping similar series before modeling.

Contribution

It proposes a novel methodology combining time series clustering with RNNs, enhancing forecasting accuracy on large, diverse datasets compared to traditional univariate methods.

Findings

Outperforms baseline LSTM in mean sMAPE accuracy

Achieves top results on CIF2016 forecasting competition dataset

Demonstrates effectiveness of clustering in heterogeneous time series forecasting

Abstract

With the advent of Big Data, nowadays in many applications databases containing large quantities of similar time series are available. Forecasting time series in these domains with traditional univariate forecasting procedures leaves great potentials for producing accurate forecasts untapped. Recurrent neural networks (RNNs), and in particular Long Short-Term Memory (LSTM) networks, have proven recently that they are able to outperform state-of-the-art univariate time series forecasting methods in this context when trained across all available time series. However, if the time series database is heterogeneous, accuracy may degenerate, so that on the way towards fully automatic forecasting methods in this space, a notion of similarity between the time series needs to be built into the methods. To this end, we present a prediction model that can be used with different types of RNN models on subgroups of similar time series, which are identified by time series clustering techniques. We assess our proposed methodology using LSTM networks, a widely popular RNN variant. Our method achieves competitive results on benchmarking datasets under competition evaluation procedures. In particular, in terms of mean sMAPE accuracy, it consistently outperforms the baseline LSTM model and outperforms all other methods on the CIF2016 forecasting competition dataset.