Developing an optimal weighting model for fund of funds portfolio construction

Abstract

This study investigates the various weighting methods of a fund of funds portfolio to determine and construct the optimal fund of funds portfolio. In most fund of funds, the underlying funds are selected with little to no regard for the risk each fund contributes to the fund of funds’ total risk level. The research tested four weighting regimes, namely one based on the Sharpe methodology, the Treynor methodology, a net asset value (NAV) momentum strategy or price momentum strategy, and finally, a momentum strategy based on equal weights. All four weighting regimes were tested over distinct time periods. The first was a 5-year period from 2011 to 2015, which was generally characterised by strong performances from most asset classes. During the second period, from 2016 to 2020, the main asset classes struggled to outperform consumer price index (CPI)+4.4%. Additionally, the study stress tested the four weighting methods under a black swan condition, namely Covid-19 in 2020. Furthermore, the study analysed how these weighting methods performed in a high inflation and later an increasing interest rate environment from 2021 to 2022. In analysing the test results, it was found that, during a growth period with low market volatility (2011–2015), all four weighting methods outperformed the South African multi asset high equity category average on an absolute and risk-adjusted basis. The equal weight portfolio delivered the second-best absolute return among the four regimes. As soon as market returns were low or slightly negative, the four weighting methods struggled to keep up with the peer group on an absolute and risk-adjusted basis. However, when risk assets were in a bear market or market correction territory, the four weighting methods performed very well, beating the peer group average on both an absolute and risk-adjusted basis. In the first quarter of 2020, equity markets entered correction territory, only to bottom on the 23rd of March 2020. The equal weight portfolio only lost 14.09% over that time, beating not only the other three weighting methods, but also the peer group average, which lost 20.66%. An interesting finding in the subsequent recovery period and inflationary period after that was that the four weighting methods underperformed the peer group average. There were two weighting methods that performed best overall. They were the price momentum and equal weighting methods. An optimal portfolio should be constructed using one of these methods. Due to the unique methodology of the price momentum strategy, it would not be possible to use both equal weighting and price momentum methodologies in one portfolio simultaneously. A portfolio manager can change the weighting method as market conditions dictate.