Invest Skeptically

In Part 1 we looked at how endpoint bias and survivor bias can be used by marketers to make their funds look better in their sales materials. Today we’ll discuss some additional information that you should look for when evaluating funds. These are examples of information that help to give you a sense of the risk-adjusted returns, or value proposition, of the manager, and, they are the type of information I wish were commonly available.

Other useful statistics for fund analysis:

• Daily/Weekly/Monthly returns to create performance statistics for rolling windows. The Zephyr Associates Inc. article that I cited in Part 1 shows a good example of a fund that looks like it is consistently outperforming the market index but when graphing rolling 36-month windows you find that the fund spent about 50% of the time under-performing!
• Daily/Weekly/Monthly peer rankings to perform the same analysis as above.
• Tracking Error — this is the standard deviation of excess (fund return minus index return for a given holding period … monthly is good) returns over many periods.
• Information Ratio — this is excess return divided by the tracking error. It is used as a “risk-adjusted” return statistic. A higher number is better.
• Sharpe Ratio — this is the fund’s excess return over the risk-free-rate divided by the standard deviation of returns of the fund. It is another measure of “risk-adjusted” return. A higher number is better.

While all the above statistics are useful, they add information to your analysis and should not be used in isolation. For example, the information and Sharpe ratios are typically not stable over time. However, a fund with a consistently high information ratio may be a good sign that the manager is highly skilled. But then again, historical returns may not be indicative of the future performance.

The sad reality is that most retail investors do not have free access to all the above information.