A very comprehensive analysis by Laurent Favre and Jose-Antonio Galeano. They start from the premise that a mean-variance framework is not optimal for evaluating the impact of hedge funds on a portfolio and develop a method based upon modified Value-at-Risk for non-normally distributed assets.
They found that constructing a portfolio without taking into account skewness and kurtosis underestimates the portfolio risk, measured with modified VaR, by 10% to 40% depending on the level of historical return.
Analyzing the distribution of hedge fund strategies returns, the average returns over the last 10 years and their correlation with a traditional portfolio, they show that the classical linear correlation and the classical linear regression cannot be relied upon with hedge funds.
Of particular interest is the finding that only three strategies, Convertible Arbitrage, Market Neutral and CTA, give diversification during market downturns. The techniques used are non-linear regressions, moving correlation and local correlation.

A wide-ranging working paper from the Yale International Centre for Finance. Authors Gary Gorton and K. Geert Rouwenhorst constructed an equally-weighted index of commodity futures over the period from July 1959 to March 2004 to study properties of commodity futures as an asset class. Among the conclusions:

• over all horizons, except monthly, the commodity futures return is negatively
  correlated to the return on the S&P 500 and long term bonds
• commodity futures are positively correlated with inflation, unexpected inflation
  and unexpected changes in inflation
• the historical return to an investment in commodity futures has far exceeded
  the return to a holder of spot commodities
• it seems that the diversification benefits of commodity futures work well when
  they are needed most. Consistent with a negative correlation, commodity
  futures earn above average returns when stocks fall below average returns

A very interesting article by William Fung, Visiting Professor of Finance, Centre of Hedge Fund Research and Education, London Business School and David A. Hsieh, Professor of Finance, Fuqua School of Business, Duke University. They apply principal component analysis to the generally accepted style groupings of fixed income hedge funds to extract common sources of risk and return. These common sources of risk are related to market risk factors, such as changes in interest rate spreads and options on interest rate spreads. They call these asset-based style factors ("ABS"). The paper finds that fixed-income hedge funds tend to be exposed to a common ABS factor:credit spreads.

The Canadian chapter of the Alternative Investment Management Association has just published its AIMA Canada Hedge Fund Primer. At 43 pages, this is an excellent introduction to hedge funds in Canada. Topics covered include the size, structure and growth of the hedge fund world, a discussion of the different types of hedge fund strategies, with very clear examples of each, risk and return. allocation considerations performance expectations and a glossary. Clear, concise, informative and easy to read...highly recommended.

A familiar study by the Center for International Securities and Derivatives Markets examining the effect of adding managed futures to a conventional stock and bond portfolio has been updated with data to December 31st, 2003.
Among the conclusions: " First, managed futures trade in markets offering investors the same integrity and safety as stock and bond markets. Second, managed futures are not more risky than traditional equity investment. Managed futures are shown on average to have a low return correlation with traditional stock and bond markets as well as with many hedge fund strategies and to offer investors the potential for reduced portfolio risk and enhanced investment return."

A concise, easy to read article by Steven Koomar of KV1 Asset Management LLC which highlights the fat tail and positive skew of the distribution of managed futures returns compared to traditional stock/bond portfolios and the associated diversification benefits.

A very instructive research paper by Carr Futures which examines the distribution and predictability of drawdowns. The researchers used Monte Carlo simulations in which they controlled for length of track record, mean return, volatility of returns, skewness, kurtosis, de-leveraging when in drawdown and survival. The only three found to be of importance were length of track record, mean return and volatility of return.

The analysis suggests that reasonable expectations can be formed about the size and frequency of drawdowns for any given investment horizon. It appears possible that for a given investment horizon and assumptions about the mean and volatility of returns, one can calculate the likelihood that a manager will experience a worse drawdown and a conditional maximum drawdown to go with it.

This paper by Bing Liang of Case Western Reserve University examines hedge funds, funds of hedge funds and CTAs' by investigating performance, risk and fund characteristics. The sample comprises 2,357 hedge funds, 597 funds of funds and 1,510 CTAs.
Among the findings are that CTAs differ from hedge funds and fund of funds in terms of trading strategies, attrition rates and survivorship bias, liquidity and correlation structures in different market environments.
"Hedge funds are highly correlated to each other and are not well hedged in down markets with liquidity squeeze.The negative correlations with other instruments make CTAs suitable hedging instruments for insuring downside risk.When adding CTAs to the hedge fund portfolio or the fund of funds portfolio, investors can benefit significantly from the risk-return trade-off."

The Canadian chapter of AIMA, the Alternative Investment Management Association, has just published this comprehensive guide. It is based upon the highly popular "Guide to Sound Practices for European Hedge Fund Managers" published last year.  Although intended for managers, any investor contemplating hedge funds will find this guide very helpful in framing due diligence and manager supervision processes.

This paper by Carol Alexander and Anca Dimitriu of the University of Reading builds a portfolio construction model designed specifically for funds of hedge funds. According to the abstract "absolute performance is targeted by selecting funds according to their abnormal return, alpha....... we find that ranking funds according to their alpha is an efficient selection process. In an extensive out-of-sample historical analysis, funds of funds that are selected this way and then allocated using constrained minimum variance optimisation are shown to perform much better than the equally weighted portfolio of all funds, or minimum variance portfolios of randomly selected funds".

This paper by Steve Kaplan and Antoinette Schoar of the University of Chicago Graduate School of Business is a ground-breaking, comprehensive analysis of returns in private equity.
The study focused on funds with largely realized returns, thus eliminating ambiguity or controversy about interim valuations. Two samples are examined, one of 746 funds and one of 1090 funds.
The abstract states: "Returns persist strongly across funds raised by individual private equity partnerships. The returns also improve with partnership experience.Better performing funds are more likely to raise follow-on funds and raise larger funds than funds that perform poorly."
The paper is highly recommended to any investor contemplating private equity.