Portfolio optimization in the presence of tail correlation

FB Abdelaziz, M Chibane - Economic Modelling, 2023 - Elsevier
Economic Modelling, 2023Elsevier
We investigate the relative performance of optimal versus naive portfolio strategies. The
accepted status on this question is that naive diversification outperforms optimal strategies.
We revisit this question using US data for equity, Treasury bonds, Gold and Crude Oil
between 2002 and 2022 by analyzing the portfolio of investors displaying constant relative
risk aversion who also consider tail behavior in the dynamics of assets. We use moment
generating functions applied to non-Gaussian processes to obtain accurate model …
Abstract
We investigate the relative performance of optimal versus naive portfolio strategies. The accepted status on this question is that naive diversification outperforms optimal strategies. We revisit this question using U.S. data for equity, Treasury bonds, Gold and Crude Oil between 2002 and 2022 by analyzing the portfolio of investors displaying constant relative risk aversion who also consider tail behavior in the dynamics of assets. We use moment generating functions applied to non-Gaussian processes to obtain accurate model estimation as well as an efficient control variate for the utility maximization problem. Our results show that risk-averse investors that are aware of tail dynamics consistently outperform the most standard portfolio strategies. In particular, highly risk-averse investors substantially outperform the so-called naive 1/N portfolio in both pre-COVID-19 and post-COVID-19 periods. Thus, true portfolio diversification requires considering both the complexity of asset dynamics and realistic risk aversion structures.
Elsevier
以上显示的是最相近的搜索结果。 查看全部搜索结果