Factor investing with risk parity portfolios

Abstract

This thesis investigates factor investing and risk parity methods by constructing seven risk parity portfolios. We find that both single-factor portfolios and multi-factor risk parity portfolios outperform the market and our benchmarks. The methods produce higher absolute returns and better risk-adjusted returns with lower volatilities and drawdowns. Therefore, the presented methodologies may benefit investors by providing more efficient portfolios and greater risk management. We use long-only factor tilt indices to construct the portfolios as this is one of the simplest ways to implement such strategies for both institutional and individual investors. Thus, our methods are both practical and realistic.

In addition, we show that there are significant diversification benefits in combining factors into multi-factor portfolios. Mainly, volatilities and drawdowns are significantly decreased when six factor tilt indices are combined using the methods described in this thesis.

Furthermore, traditional diversification methods can lead to skewed portfolio risk profiles where equity risk is significantly overweighted. Risk-based diversification methods offer an alternative approach where diversification is based on the risk contribution of each asset. This results in more balanced portfolio risk profiles. Therefore, we also analyze the compositions of the constructed portfolios and find that the majority overweight low volatility assets.