Portfolio optimisation is still an effective risk management tool despite a growing notion that an evenly weighted portfolio presents a better risk alternative, State Street Associates senior managing director Sebastien Page said.
Page’s criticism of the belief that equally weighted portfolios outperform optimised portfolios centres on the research supporting the theory that uses historical rolling five year returns to forecast future returns.
To test the theory that optimised portfolios are the better performers State Street performed an empirical study covering 13 different data sets.
“We used data going back as far as 1926 and constructed over 50,000 optimised portfolios and measured their performance out of sample,” Page said.
“We covered asset allocation, allocation to risk premiums, and alpha. So we’ll go and choose security selection, allocations across commodity contracts, allocating across hedge fund styles, and allocating across active managers,” he said.
“In all cases we found even simple optimisation, with the assumption that you don’t have any forecasting skill, but you simply either build a risk minimising portfolio or a portfolio that is consistent with reasonable risk premiums, you get the result that optimisation improves Sharpe ratios by anywhere between 20 to 60 per cent on average.”
The train of thought that equally weighted portfolios may have gained weight as a result of the global financial crisis as it offers simplicity and transparency, two factors investors are demanding, according to Page.
“It is transparent and people might have the perception that to equal weight your portfolio is going the safe route but it can be demonstrated very easily that in a lot of cases that it isn’t and those portfolios are riskier,” he said.
