A Hypothesis on Good Practices for AI-based Systems for Financial Time Series Forecasting: Towards Domain-Driven XAI Methods

Paper: arXiv 2311.07513

Authors: Branka Hadji Misheva, Joerg Osterrieder

Abstract

Machine learning and deep learning have become increasingly prevalent in financial prediction and forecasting tasks, offering advantages such as enhanced customer experience, democratising financial services, improving consumer protection, and enhancing risk management. However, these complex models often lack transparency and interpretability, making them challenging to use in sensitive domains like finance. This has led to the rise of eXplainable Artificial Intelligence (XAI) methods aimed at creating models that are easily understood by humans. Classical XAI methods, such as LIME and SHAP, have been developed to provide explanations for complex models. While these methods have made significant contributions, they also have limitations, including computational complexity, inherent model bias, sensitivity to data sampling, and challenges in dealing with feature dependence. In this context, this paper explores good practices for deploying explainability in AI-based systems for finance, emphasising the importance of data quality, audience-specific methods, consideration of data properties, and the stability of explanations. These practices aim to address the unique challenges and requirements of the financial industry and guide the development of effective XAI tools.

Complexity vs Empirical Score

• Math Complexity: 2.5/10
• Empirical Rigor: 3.0/10
• Quadrant: Philosophers — conceptual discussion, limited math and data

Why this score: The paper provides a high-level conceptual framework and set of best practices for Explainable AI (XAI) in finance. However, it lacks dense mathematical formulation, rigorous backtesting, and quantitative empirical validation.

Research Flowchart

  flowchart TD
    A[Research Goal: Establish Good Practices for XAI in Financial Time Series Forecasting] --> B[Data Inputs: Financial Time Series & Cross-Asset Market Data]
    B --> C[Methodology: Evaluate Classical XAI Methods LIME & SHAP]
    C --> D[Computational Analysis: Assess Complexity, Stability, Bias & Feature Dependence]
    D --> E[Outcomes: Framework for Domain-Driven XAI & Audience-Specific Guidelines]

Cited References (1)

1. Feature weighting : The coefficients or weights derived from the interpretable model correspond to the influence or significance of each feature in the local decision boundary

0 of 1 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.