I was rereading an article by CXO Advisory on rating guru predictions. This is an important study because it bursts the bubble that the investment gurus have some kind of magic power at predicting. The evidence suggests they are worse than a flip of a coin. This study graded close to 7000 forecasts from some big names in the industry. These discretionary managers do not have any special insights.
I am not indicting these experts. Their investment processes work for them. Their good picks may way outstrip their loses. The study was carefully conducted and the authors are careful with their conclusions, but the results are sobering. If you cannot trust the experts to be better than a flip of the coin, what is the next best solution?
There are two answers to the problem of poor forecasting skill. One, diversify. Hold a basket of many alternatives that are truly uncorrelated. Two, use prices as your guide. The expert can be the weighted opinion of the market itself. In the simplest form, the best guess of tomorrow may be what has happened today. (One of the key actions of any trend-follower is filtering data so that some of the noise of any trading day is diminished.)
Actually the two concepts are tied together because diversification is connected to price behavior through the measures volatility and correlation. Prices may not be a perfect guide and there can be deviations from fair value, but using price data as base predictions is always a good starting point. If you do not have an information advantage or a processing advantage, a price-based system is the bets place to start.
No comments:
Post a Comment