Hierarchical Risk Parity and Minimum Variance Portfolio Design on NIFTY 50 Stocks

TL;DR

This paper compares hierarchical risk parity and minimum variance portfolio algorithms applied to NIFTY 50 stocks, showing HRP's better out-of-sample performance despite CLA's superior training results.

Contribution

It introduces a systematic approach using hierarchical risk parity and critical line algorithms for portfolio design in the Indian stock market.

Findings

HRP outperforms CLA on test data

CLA performs better on training data

Portfolio performance varies between algorithms

Abstract

Portfolio design and optimization have been always an area of research that has attracted a lot of attention from researchers from the finance domain. Designing an optimum portfolio is a complex task since it involves accurate forecasting of future stock returns and risks and making a suitable tradeoff between them. This paper proposes a systematic approach to designing portfolios using two algorithms, the critical line algorithm, and the hierarchical risk parity algorithm on eight sectors of the Indian stock market. While the portfolios are designed using the stock price data from Jan 1, 2016, to Dec 31, 2020, they are tested on the data from Jan 1, 2021, to Aug 26, 2021. The backtesting results of the portfolios indicate while the performance of the CLA algorithm is superior on the training data, the HRP algorithm has outperformed the CLA algorithm on the test data.