Robust Monitoring of Time Series with Application to Fraud Detection

TL;DR

This paper introduces a robust framework for detecting outliers and structural changes in short, seasonal time series, crucial for identifying fraud in transaction data.

Contribution

It combines robust regression with alternating least squares and introduces the double wedge plot for effective visualization of anomalies.

Findings

Effective detection of outliers and level shifts in real-world fraud detection scenarios.

The methodology successfully identified suspicious transactions in EU import data.

The approach improves robustness over traditional methods.

Abstract

Time series often contain outliers and level shifts or structural changes. These unexpected events are of the utmost importance in fraud detection, as they may pinpoint suspicious transactions. The presence of such unusual events can easily mislead conventional time series analysis and yield erroneous conclusions. In this paper we provide a unified framework for detecting outliers and level shifts in short time series that may have a seasonal pattern. The approach combines ideas from the FastLTS algorithm for robust regression with alternating least squares. The double wedge plot is proposed, a graphical display which indicates outliers and potential level shifts. The methodology was developed to detect potential fraud cases in time series of imports into the European Union, and is illustrated on two such series.