Are VaR models effective in capturing downside risk in alternative investment funds? Insights from a cross-country study

Abstract

In this paper, we analyze the downside risk of alternative investment funds (AIFs) in a cross-country setting over a period of 2015–2021, using popular Value-at-Risk (VaR henceforth) models, and test the efficacy of such models in capturing the volatility posed by these funds. We estimate VaR in the presence of three error distributions, including normal distribution, Student’s t distribution, and GED distribution. Using weekly return data of 991 AIFs from 28 countries, over the period 2015–2021, we find that most of the funds, irrespective of country representation, exhibited a significant proportion of their weekly returns to be negative. To statistically validate our findings we use three backtesting approaches, i.e., Jorion’s failure rate, Kupiec’s proportion of failure (POF) tests and Christoffersen’s independence test. Our findings indicate the presence of significant downside risk for AIFs, which these popular VaR models are unable to capture. These findings highlight the need for investors and fund managers to be cautious when relying solely on popular VaR models to manage downside risk in AIF and suggest that further research is needed to develop more effective risk management strategies for AIF, particularly in light of their complex structure and asymmetric return distributions.