Can Investors Profit from Information Diveristy? The Wisdom of Crowds in Security Analyst Recommendations

Abstract

There is heterogeneity in individual forecasts of any variable — inflation, corporate earnings, etc. The standard consensus estimate takes a simple average of individual forecasts, implicitly treating each forecast as a common signal plus noise. If some individuals know more than others, then a consensus estimate is not necessarily the optimal way to combine forecasts. I show how a recently developed statistical technique can infer overlap in information across agents and I apply it to stock recommendations of sell-side analysts. I find a trading strategy that delivers an alpha of 2-3% on an annualized basis, net of transaction costs, suggesting that information diversity is prevalent, economically significant, and tradable.