As we noted last week, for mutual fund analysis peer-group construction often plays a key role. Many methods for constructing peer groups exist. In this article we focus on a way to construct peer groups for Lipper’s Absolute Return funds classification. The factors we use are: performance, management fees, advisor fees, and total expenses. (These are the same factors we used last week to construct peer groups for Large-Cap Core Equity [LCCE] funds.) The universe is composed of those Absolute Return funds that have annualized returns for 2010–2012 as well as having advisor fees and management fees. They must also have reported a total expense ratio. The expense data are as of year-end 2012.
We follow the same dissimilarity indices methodology to compute distances between funds as we did last week and refer the reader to the same document for a brief introduction. And, as we did last week, we look at peers for the largest (by assets under management [AUM]) Absolute Return fund: Absolute Strategies Fund: Institutional (ticker symbol ASFIX).
In the table below we show the closest funds to ASFIX:
Table 1. Peer Group for ASFIX
ASFIX is the fund marked in green. The funds in blue are the best “fit” peer group for ASFIX, while those in tan are close companions (you could call them “cousins” of ASFIX). It bears noting that all the remaining 13 Absolute Return funds (primary share class only) are a significant distance from ASFIX and would not be the best funds to include in this peer group.
Since peer-group construction often relies heavily on returns, the author notes that in this case it would be very difficult to use any one-, two-, or three-year annualized returns to pick peer funds. The volatility of each fund’s annual returns would thwart the best of the single-factor methodologies.
We show below the “collapse” of our six-dimensional, i.e., six-factor, distance measures. They are “dropped” or folded down to three dimensions:
Figure 1. The Spread or Distance Between Absolute Return Funds
Each pink dot represents one of the 21 funds. As the reader can see, there are two significant outliers in the list of funds. These are marked in yellow in Table 1: Newmark Risk Managed Opportunistic Fund and Absolute Opportunities Fund. Forming peers groups for these two funds would be very difficult, given the great distance they are from the other funds. However, even the remaining funds are a mix of close and not really close. This probably means that at best the peer group for the “inner funds” would not number any higher than it does for ASFIX and could very well be smaller.
In this article we have shown an alternative way to construct peer groups across multiple factors simultaneously: computing distances between each of the funds.