D\'ej\`a vu: A data-centric forecasting approach through time series cross-similarity

TL;DR

This paper introduces a data-centric, model-free forecasting method called 'forecasting with similarity' that leverages cross-series pattern matching to improve forecast accuracy, especially for short time series.

Contribution

It proposes a novel cross-similarity approach for time series forecasting, moving beyond traditional self-similarity and model assumptions, enabling effective forecasting with limited data.

Findings

Competitive accuracy in point forecasts

Effective prediction intervals

Applicable to short time series

Abstract

Accurate forecasts are vital for supporting the decisions of modern companies. Forecasters typically select the most appropriate statistical model for each time series. However, statistical models usually presume some data generation process while making strong assumptions about the errors. In this paper, we present a novel data-centric approach -- `forecasting with similarity', which tackles model uncertainty in a model-free manner. Existing similarity-based methods focus on identifying similar patterns within the series, i.e., `self-similarity'. In contrast, we propose searching for similar patterns from a reference set, i.e., `cross-similarity'. Instead of extrapolating, the future paths of the similar series are aggregated to obtain the forecasts of the target series. Building on the cross-learning concept, our approach allows the application of similarity-based forecasting on series with limited lengths. We evaluate the approach using a rich collection of real data and show that it yields competitive accuracy in both points forecasts and prediction intervals.