Data Augmentation in Time Series Forecasting through Inverted Framework

TL;DR

This paper introduces DAIF, a novel real-time data augmentation method tailored for the inverted framework in multivariate time series forecasting, improving model robustness and accuracy.

Contribution

It proposes the first real-time augmentation techniques, Frequency Filtering and Cross-variation Patching, specifically designed for the inverted framework in MTS forecasting.

Findings

DAIF improves forecasting accuracy across multiple datasets.

DAIF effectively mitigates noise and preserves temporal dependencies.

Experiments validate the superiority of DAIF over existing methods.

Abstract

Currently, iTransformer is one of the most popular and effective models for multivariate time series (MTS) forecasting. Thanks to its inverted framework, iTransformer effectively captures multivariate correlation. However, the inverted framework still has some limitations. It diminishes temporal interdependency information, and introduces noise in cases of nonsignificant variable correlation. To address these limitations, we introduce a novel data augmentation method on inverted framework, called DAIF. Unlike previous data augmentation methods, DAIF stands out as the first real-time augmentation specifically designed for the inverted framework in MTS forecasting. We first define the structure of the inverted sequence-to-sequence framework, then propose two different DAIF strategies, Frequency Filtering and Cross-variation Patching to address the existing challenges of the inverted framework. Experiments across multiple datasets and inverted models have demonstrated the effectiveness of our DAIF.