Mutual/Hedge Funds

Does the expanding set of exchange-traded funds (ETF) support reliable replication (cloning) of future returns for some hedge funds? In their December 2014 paper entitled  “Smart Beta ETF Portfolios: Cloning Beta Active Hedge Funds”, Jun Duanmu, Yongjia Li and Alexey Malakhov test replication of top risk factor-driven (beta-active) hedge funds using portfolios of ETFs. The selected hedge funds perform well historically and are especially suited to cloning because of their dependence on known risk factors. The hedge fund selection and cloning process involves repeating four steps annually based on two years of monthly historical data. Specifically, each year the authors:

1. Identify the fourth of hedge funds with returns most strongly correlated with known risk factors.
2. Iterate cluster analysis 100 times to identify ETFs most representative (highest correlation of monthly returns with the mean return of the cluster) of up to 100 clusters to serve as risk factor proxies.
3. Use an optimization tool on each of the 100 cluster analyses to combine representative ETFs into 100 clone models of pre-fee (risk factor perspective) monthly returns for each target hedge fund.
4. Apply the Bayesian information criterion (which addresses data snooping bias via a penalty for model complexity) to select the best clone model for each target hedge fund.

They then test, starting in 2005 (when enough historical ETF data become available), the ability of winning clone models to match post-fee (investor perspective) monthly returns of target hedge funds for one year out-of-sample. They mitigate backfill bias in hedge fund returns (only funds with good starts begin publicizing their returns) by excluding the first 24 months of reported returns. They suppress survivorship bias by including funds that later stop reporting. Using monthly net returns for 2,014 hedge funds (963 live and 1,051 dead) during 1994 through 2012 and monthly returns for 1,313 passive ETFs as available during 1997 through 2012, they find that: Keep Reading

Do predictable monthly outflows from and inflows to mutual funds drive the Turn-of-the-Month (TOTM) effect, a concentration of positive stock market returns around the turns of calendar months? In their November 2014 paper entitled “Dash for Cash: Month-End Liquidity Needs and the Predictability of Stock Returns”, Kalle Rinne, Matti Suominen and Lauri Vaittinen explore TOTM with focus on the effects of: (1) month-end flows from mutual funds to retirees and dividend-collecting investors; and, (2) beginning-of-month flows from working investors to mutual funds. To account for trade settlement rules, funds must sell stocks at least three trading days before the end of the month to raise cash for expected month-end outflows. The authors therefore define a TOTM interval from three trading days before through three trading days after the last trading day of the month. They also consider intervals of five trading days before TOTM to measure the effect of fund selling and five trading days after TOTM  to measure reversion from fund buying. Using daily value-weighted, (mostly) total return stock market indexes for the U.S. since 1926 and for 24 other developed markets as available during January 1980 through January 2014, and data for individual U.S. stocks and mutual funds during January 1980 through December 2013, they find that: Keep Reading

Do hedge fund-like mutual funds work like hedge funds? In his September 2014 paper entitled “Hedge Funds versus Mutual Funds (2): An Examination of Multialternative Mutual Funds”, David McCarthy evaluates mutual funds categorized by Morningstar as “multialternative” after further subcategorizing them as: Global Asset Allocation (active asset allocation across a broad set of global markets); Multistrategy (investing across distinct investment styles); and, Replication (quantitatively mimicking the returns of a hedge fund index). He profiles these groups, compares their asset class and factor exposures to those of hedge fund indexes, and compares their performances to those of hedge fund indexes. He considers three benchmarks: the Hedge Fund Research (HFR) Relative Value Multistrategy Index; the HFR Fund of Funds Diversified Index; and, the long-only Morningstar Global Allocation Index. Using monthly returns as available for 30 Global Asset Allocation, 33 Multistrategy and four Replication mutual funds established as of January 2013 along with contemporaneous returns for asset class proxies, factors and benchmarks during January 2008 through December 2013, he finds that: Keep Reading

How successful are active equity mutual fund managers in timing their domestic markets worldwide? In their August 2014 paper entitled “Market Timing Around the World”, Javier Vidal-Garcia, Marta Vidal and Duc Khuong Nguyen employ daily returns to measure the effectiveness of mutual fund market exposure adjustments made more frequently than monthly. They also examine fund timing performance under different economic conditions. Their fund universe consists of 8,680 actively managed, open-end, diversified, domestic live and dead equity mutual funds registered in 35 countries (about 69% are U.S.-registered). Using daily total returns in local currencies and characteristics for these funds, along with contemporaneous country economic data, during January 1990 through December 2013, they find that: Keep Reading

How should investors use Morningstar mutual fund ratings/grades to select mutual funds? In his July 2014 paper entitled “Morningstar Mutual Fund Measures and Selection Model”, John Haslem surveys the five kinds of Morningstar mutual fund ratings and grades: (1) Morningstar star ratings (one to five stars); (2) analyst ratings (gold, silver, bronze, neutral and negative); (3) total pillar ratings (positive, neutral or negative for fund people, process, parent, performance and price); (4) upside/downside capture ratios; and, (5) stewardship ratings (culture, incentives, fees, board quality and regulatory history). Based on the body of research about the predictive power of Morningstar ratings/grades, he chooses three criteria for screening mutual funds:

1. Star rating of 4 or 5 and analyst rating of gold or silver.
2. Upside capture ratios greater than downside capture ratios for all three of 3-year, 5-year and 10-year past performance intervals.
3. Total stewardship grade of A.

He applies these criteria to the set of Vanguard actively managed diversified (not sector) U.S. equity mutual funds. His selections are current winners, with empirical testing requiring future performance data. Applying the chosen criteria to the specified set of Vanguard funds (about 20 funds), he finds that: Keep Reading

Are widely used volatility-adjusted investment performance metrics, such as Sharpe ratio, robust to different measurement intervals? In the July 2014 version of their paper entitled “The Divergence of High- and Low-Frequency Estimation: Implications for Performance Measurement”, William Kinlaw, Mark Kritzman and David Turkington examine the sensitivity of such metrics to the length of the return interval used to measure it. They consider hedge fund performance, conventionally estimated as Sharpe ratio calculated from monthly returns and annualized by multiplying by the square root of 12. They also consider mutual fund performance, usually evaluated as excess return divided by excess volatility relative to an appropriate benchmark (information ratio). Finally, they consider Sharpe ratios of risk parity strategies, which periodically rebalance portfolio asset weights according to the inverse of their return standard deviations. Using monthly and longer-interval return data over available sample periods for each case, they find that: Keep Reading

Is there a best way to exploit unusual closed-end fund discounts to net asset value? In their July 2014 paper entitled “Exploiting Closed-End Fund Discounts: The Market May Be Much More Inefficient Than You Thought”, Dilip Patro, Louis Piccotti and Yangru Wu construct two regression models to predict closed-end fund returns:

1. One model is a simple regression based on the past relationship between monthly fund discount and next-month fund return.
2. The other augments the first by including a term that accounts for effects of changes in the discount over a recent (optimized) interval.

They test whether these models beat a naive strategy that trades only on current closed-end fund discounts. They focus on Sharpe ratio as a key performance metric. Using monthly prices, net asset values and classifications for 377 U.S. closed-end funds as available during August 1984 through December 2011 and contemporaneous monthly four-factor (market, size, book-to-market, momentum) and liquidity factor returns, they find that: Keep Reading

“Mutual Fund Hot Hand Performance” tests a “hot hand” strategy that each year picks the top performer from the Vanguard family of diversified equity mutual funds (not including sector funds) and holds that winner the next year. A subscriber suggested a robustness test using the Fidelity family of diversified equity mutual funds. To support the test, we select all Fidelity diversified U.S. and international equity mutual funds that bear no transaction fee, are open to new investors and have a history of at least three years. We consider the total return on the S&P 500 Index (with dividends estimated from Robert Shiller’s data) and SPDR S&P 500 (SPY) as benchmarks. As in the prior analysis of Vanguard funds, we pick end of June to end of the next June for annual return measurement intervals. To simplify analysis, we assume the “hot hand” mutual fund on the next-to-last trading day of June is the same as that for the end of June. We assume that there are no costs or holding period constraints/delay for switching from one fund to another. Using annual returns for the S&P 500 Index plus Shiller’s dividend data and annual returns for SPY and Fidelity diversified equity mutual funds as available from Yahoo!Finance during June 1980 through June 2014, we find that: Keep Reading

How do hedge funds electing not to report to a commercial database differ from those that do? In their July 2014 paper entitled “What Happens ‘Before the Birth’ and ‘After the Death’ of a Hedge Fund?”, Vikas Agarwal, Vyacheslav Fos and Wei Jiang compare performances of equity hedge funds before they begin self-reporting, while they are self-reporting; and after they stop self-reporting to commercial databases. They develop a sample of hedge funds that do and do not self-report by matching hedge fund Securities and Exchange Commission (SEC) Form 13F filings to listings of hedge funds that self-report to any of five major hedge fund commercial databases. They then identify subsamples of hedge funds that: (1) initially do not but later do self-report; and, (2) initially do but later do not self-report. They then use the long-only equity holdings in series of Form 13F to analyze performances and characteristics within subsamples. Using 1,199 series of Form 13Fs for firms that are clearly hedge funds during 1980 through 2008 and contemporaneous data for hedge funds self-reporting to commercial databases, they find that: Keep Reading

Are the sources of active mutual fund risk mostly common (systematic) or unique (idiosyncratic)? In his July 2014 paper entitled “Components of Portfolio Variance: R2, SelectionShare and TimingShare”, Anders Ekholm decomposes mutual fund return variance (risk) into three sources: (1) passive systematic factor exposure (R-squared); (2) active security selection or stock picking (SelectionShare); and, (3) active systematic factor timing (TimingShare). He demonstrates estimation of these three components based on mutual fund returns (reflecting daily manager actions) rather than holdings (known only via quarterly snapshots). He employs the widely used four-factor (market, size, book-to-market, momentum) model of stock returns to define systematic risk. Using daily returns for a broad sample of actively managed U.S. equity mutual funds and for the four factors during 2000 through 2013, he finds that: Keep Reading