# Is being `Robust' beneficial?: A perspective from the Indian market

## TL;DR

This paper extends robust optimization to downside risk measures like VaR and CVaR, empirically demonstrating their superior performance over traditional methods in the Indian market, especially with larger portfolios.

## Contribution

It introduces robust optimization frameworks for VaR and CVaR, and empirically evaluates their effectiveness using market and simulated data.

## Key findings

- Robust models outperform base models in larger stock portfolios.
- Robust VaR and CVaR models show improved risk minimization.
- Empirical results support the practical usefulness of robust approaches.

## Abstract

The problem of data uncertainty has motivated the incorporation of robust optimization in various arenas, beyond the Markowitz portfolio optimization. This work presents the extension of the robust optimization framework for the minimization of downside risk measures, such as Value-at-Risk (VaR) and Conditional Value-at-Risk (CVaR). We perform an empirical study of VaR and CVaR frameworks, with respect to their robust counterparts, namely, Worst-Case VaR and Worst-Case CVaR, using the market data as well as the simulated data. After discussing the practical usefulness of the robust optimization approaches from various standpoints, we infer various takeaways. The robust models in the case of VaR and CVaR minimization exhibit superior performance with respect to their base versions in the cases involving higher number of stocks and simulated setup respectively.