Paper: arXiv 2507.02011

Authors: Vidya Sagar G, Shifat Ali, Siddhartha P. Chakrabarty

Abstract

This paper presents a machine learning driven framework for sectoral stress testing in the Indian financial market, focusing on financial services, information technology, energy, consumer goods, and pharmaceuticals. Initially, we address the limitations observed in conventional stress testing through dimensionality reduction and latent factor modeling via Principal Component Analysis and Autoencoders. Building on this, we extend the methodology using Variational Autoencoders, which introduces a probabilistic structure to the latent space. This enables Monte Carlo-based scenario generation, allowing for more nuanced, distribution-aware simulation of stressed market conditions. The proposed framework captures complex non-linear dependencies and supports risk estimation through Value-at-Risk and Expected Shortfall. Together, these pipelines demonstrate the potential of Machine Learning approaches to improve the flexibility, robustness, and realism of financial stress testing.

Complexity vs Empirical Score

  • Math Complexity: 8.5/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper utilizes advanced machine learning concepts like Variational Autoencoders and probabilistic modeling, indicating high mathematical complexity, while providing a concrete implementation pipeline with specific risk metrics (VaR, ES) and rolling-window methodology, demonstrating strong empirical rigor.

Research Flowchart

  flowchart TD
  A["Research Goal: Develop ML Framework<br>for Indian Financial Sector Stress Testing"] --> B["Data: Sectoral Equity Indices<br>for 5 Key Sectors"]
  
  B --> C{"ML Architecture"}
  
  C --> D["Pipeline 1: PCA & AE<br>Dimensionality Reduction"]
  C --> E["Pipeline 2: VAE<br>Probabilistic Latent Space"]
  
  D --> F["Scenario Generation"]
  E --> G["Monte Carlo Simulation<br>from VAE Latent Distribution"]
  
  F --> H["Outcome: Risk Metrics<br>VaR & Expected Shortfall"]
  G --> H
  
  style A fill:#f9f,stroke:#333,stroke-width:2px
  style H fill:#bbf,stroke:#333,stroke-width:2px

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a machine learning driven framework for sectoral stress testing in the Indian financial market using PCA, autoencoders and variational autoencoders.

Source sentence
This paper presents a machine learning driven framework for sectoral stress testing in the Indian financial market, focusing on financial services, information technology, energy, consumer goods, and pharmaceuticals.

Assumes 1

  • Return series are stationary, making them suitable for PCA which assumes stationarity. statistical
    Source sentence
    The test results yield low p-values across all stocks, providing strong statistical evidence to reject the null hypothesis of a unit root. This indicates that the return series are stationary, making

Data

  • Historical daily closing prices of Indian equity sector stocks · Yahoo Finance market-prices public
    2004-2024 · Five sectors (Financial Services, IT, Consumer Goods, Energy, Pharmaceuticals) with five c
    Source sentence
    The dataset comprises daily closing prices collected over a 20-year period (2004–2024), retrieved from Yahoo Finance using the yfinance Python library.

Recognised sources: Yahoo Finance

Costs

Not applicable The paper reports no trading or execution result.

Finds 2

  • Consistent sector-level vulnerabilities are observed across all three pipelines, especially in Financial Services and Energy during periods of elevated market stress. statistical
    Source sentence
    Our results show consistent sector-level vulnerabilities across all three pipelines, particularly in Financial Services and Energy during periods of elevated market stress.
  • The framework improves risk sensitivity and interpretability while providing a flexible tool for forward‑looking stress testing. other
    Source sentence
    The resultant framework improves risk sensitivity and interpretability, while offering a flexible tool for forward-looking stress testing.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 12 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (7)

  1. Variational Autoencoders (VAEs) – Simply Explained (2021)
  2. Principal component analysis (1987) — link
  3. Autoencoders (2021) — link
  4. An Introduction to Variational Autoencoders (2019) — link
  5. Autoencoder Method—Wolfram Language Documentation (2024)
  6. Principles for Sound Stress Testing Practices and Supervision (2009) — link
  7. Financial Stability Report – June 2023 (2023) — link

0 of 7 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.