Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Financial Anomaly Detection for the Canadian Market

Luigi Caputi, Nicholas Meadows

Abstract

In this work we evaluate the performance of three classes of methods for detecting financial anomalies: topological data analysis (TDA), principal component analyis (PCA), and Neural Network-based approaches. We apply these methods to the TSX-60 data to identify major financial stress events in the Canadian stock market. We show how neural network-based methods (such as GlocalKD and One-Shot GIN(E)) and TDA methods achieve the strongest performance. The effectiveness of TDA in detecting financial anomalies suggests that global topological properties are meaningful in distinguishing financial stress events.

Financial Anomaly Detection for the Canadian Market

Abstract

In this work we evaluate the performance of three classes of methods for detecting financial anomalies: topological data analysis (TDA), principal component analyis (PCA), and Neural Network-based approaches. We apply these methods to the TSX-60 data to identify major financial stress events in the Canadian stock market. We show how neural network-based methods (such as GlocalKD and One-Shot GIN(E)) and TDA methods achieve the strongest performance. The effectiveness of TDA in detecting financial anomalies suggests that global topological properties are meaningful in distinguishing financial stress events.

Paper Structure

This paper contains 19 sections, 12 equations, 5 figures.

Table of Contents

  1. Introduction
  2. Methodology
  3. Data
  4. Networks construction
  5. Graph Neural Networks:
  6. One-Shot GNN
  7. Knowledge Distillation GNN (Glocal KD (GINE))
  8. Principal Component Analysis
  9. Topological Data Analysis
  10. Anomaly Detection
  11. Anomaly Scoring
  12. Experiments and Results
  13. Experiment Description
  14. Performance Metrics
  15. Software
  16. ...and 4 more sections

Figures (5)

  • Figure 1: Pipeline of the Main Analysis
  • Figure 2: Scores for Different Anomaly Detection Methods (TSX-60)
  • Figure 4: Scores for Different Anomaly Detection Methods (DJIA)
  • Figure : Legend: 1-Start of Mortgage Crisis, 2-Peak of 2008 Crisis, 3-Greek Debt Crisis, 4- Taper Tanrum, 5 - Oil $< \$40$, 6 - Oil $< \$20$, 7- COVID19
  • Figure : Legend: 1 - Start of Mortgage Crisis, 2 - 2008 Financial Crisis , 3 - Flash Crash, 4 - Greek Debt Crisis, 5 - 2015 Stock Market Selloff, 6 - 2018 Stock Market Correction, 7 - COVID19