Long-Term Outlier Prediction Through Outlier Score Modeling

TL;DR

This paper introduces a novel unsupervised framework for long-term outlier prediction in time series, enabling forecasting of outlier likelihoods far into the future beyond immediate detection.

Contribution

It proposes a simple, model-independent two-layer method that combines standard outlier detection with temporal outlier score prediction for long-term forecasting.

Findings

Performs well on synthetic datasets in detection and prediction tasks.

Serves as a strong baseline for future outlier forecasting research.

Addresses a key gap in long-term outlier prediction capabilities.

Abstract

This study addresses an important gap in time series outlier detection by proposing a novel problem setting: long-term outlier prediction. Conventional methods primarily focus on immediate detection by identifying deviations from normal patterns. As a result, their applicability is limited when forecasting outlier events far into the future. To overcome this limitation, we propose a simple and unsupervised two-layer method that is independent of specific models. The first layer performs standard outlier detection, and the second layer predicts future outlier scores based on the temporal structure of previously observed outliers. This framework enables not only pointwise detection but also long-term forecasting of outlier likelihoods. Experiments on synthetic datasets show that the proposed method performs well in both detection and prediction tasks. These findings suggest that the method can serve as a strong baseline for future work in outlier detection and forecasting.