Abstract This study examines whether the inclusion of cryptoassets improves optimal portfolio performance for key investor archetypes in behavioral finance: Cumulative Prospect Theory (CPT), Markowitz, and Loss Averse investors. We develop a framework that integrates Second-Order Stochastic Dominance (SSD) with Stochastic Spanning to construct optimal portfolios, accounting for investors’ subjective risk perceptions and narrow framing. Our methodology incorporates non-stationarities, asset return bubbles, and the safe-haven role of gold while also considering the impact of the COVID-19 pandemic. Empirical results, based on a rolling-window analysis of business-day returns, indicate that while the traditional portfolio universe (stocks, bonds, and gold) generally spans the augmented universe (including cryptoassets) in-sample, out-of-sample tests reveal that cryptoassets enhance performance, particularly for CPT and Loss Averse investors. The findings suggest that behavioral criteria do not overturn the fundamental instability and risk characteristics of crypto assets, although under certain behavioral biases, investor risk perceptions shape portfolio choices in ways that cryptoassets can appear as attractive investment options.
Arvanitis et al. (Mon,) studied this question.