Encoded Value-at-Risk: A Predictive Machine for Financial Risk Management

TL;DR

This paper introduces Encoded VaR, a novel neural network-based method using Variational Auto-encoders to measure market risk, capable of reproducing market scenarios and learning dependencies without distribution assumptions.

Contribution

The paper presents a new AI-driven approach for financial risk measurement that outperforms or matches existing methods in out-of-sample tests.

Findings

Encoded VaR effectively reproduces market scenarios.

It increases the signal-to-noise ratio in financial data.

It performs competitively with established VaR algorithms.

Abstract

Measuring risk is at the center of modern financial risk management. As the world economy is becoming more complex and standard modeling assumptions are violated, the advanced artificial intelligence solutions may provide the right tools to analyze the global market. In this paper, we provide a novel approach for measuring market risk called Encoded Value-at-Risk (Encoded VaR), which is based on a type of artificial neural network, called Variational Auto-encoders (VAEs). Encoded VaR is a generative model which can be used to reproduce market scenarios from a range of historical cross-sectional stock returns, while increasing the signal-to-noise ratio present in the financial data, and learning the dependency structure of the market without any assumptions about the joint distribution of stock returns. We compare Encoded VaR out-of-sample results with eleven other methods and show that it is competitive to many other well-known VaR algorithms presented in the literature.