Machine Learning Based Stress Testing Framework for Indian Financial Market Portfolios

TL;DR

This paper introduces a machine learning framework for sectoral stress testing in the Indian financial market, utilizing advanced autoencoder techniques to generate realistic stress scenarios and improve risk assessment accuracy.

Contribution

It develops a novel stress testing methodology combining PCA, Autoencoders, and Variational Autoencoders for probabilistic scenario generation in financial risk analysis.

Findings

Enhanced scenario realism through probabilistic modeling

Improved risk estimation with Value-at-Risk and Expected Shortfall

Demonstrated robustness over traditional stress testing methods

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.