Monday, 24 August 2009

An Innovative Asset Allocation Framework for Alternative Investment Strategies

Slide 5I have recently developed a dynamic asset allocation model that could be used to address the need of investors focusing on absolute return performance, as compared to those investors with relative return performance mandate. Basically, the model is developed to try to address the following issues associated with traditional asset allocation models.

Firstly, portfolio optimization tools based on normally distributed as
set returns (Markowitz) no longer give valid outcomes, as the inclusion of alternative investments introduces skewness and kurtosis (fat tails) to the probability distribution of the returns of a combined portfolio. Risk measurement tools may underestimate the true risk-characteristics of a portfolio. One needs a distribution with adjustable skewness and kurtosis, which gives a higher probability to outliers than the normal distribution.

Secondly, given the non-normality of the return distribution
, the traditional risk measure - standard deviation - is unable to address the higher probability of extreme losses (fat tails) faced by most investors. An intuitive measure of risk (short fall risk) is used to only penalize downside risk in the portfolio risk budgeting.

Thirdly, traditional portfolio optimization with asset weights constraints might not generate truly risk-diversified portfolio. The resulting optimal portfolios tend to be overly concentrated in a very limited subset of the full assets or securities spectru
m. Traditional 60/40 (i.e., S&P 500 and Lehman Aggregate) or so-called balanced portfolios do not offer investors true diversification because the 60% stock allocation (S&P 500) actually accounts for almost 95% of the portfolio risk. In a sense, 60/40 portfolios put almost all the “eggs” in one basket. When (not if) the stock market has a severe downturn (as witnessed recently), 60/40 portfolios would also suffer tremendous losses.

The following example shows the risk profile of an optimal portfolio, compared to its three underlying component strategies. The component strategies are long/short (market neutral) strategies in the equities and commodities space.