Computer Science · Artificial Intelligence
Comments welcomed This paper subsumes two earlier manuscripts titled “Stock Characteristics and Analyst Stock Recommendations ” and “The Information Content of Analyst Stock Recommendations. ” We have
Narasimhan Jegadeesh, Joonghyuk Kim, Susan D. Krische, Charles M. C. Lee, Louis Chan, Tom Dyckman, Paul Irvine, Josef Lakonishok, Michael Weisbach
See on the mapAbstract
Quantitative Research Conference. We would also like to thank Zacks Investment Research for providing the analyst recommendation data and I/B/E/S for providing the analyst earnings forecasts. First Call Corporation als o provided us with recommendations data used in one of our earlier manuscripts. 2 Analyzing the analysts: When do recommendations add value? We show that financial analysts from sell-side firms generally recommend “glamour ” (i.e., positive momentum, high growth, high volume, and relatively expensive) stocks. Naïve adherence to these recommendations can be costly, because the level of the consensus recommendation adds value only among stocks with positive quantitative characteristics (i.e., high value and positive momentum stocks). Among stocks with negative quantitative characteristics, higher consensus recommendations are associated with worse subsequent returns. In contrast, the quarterly change in the consensus recommendation is a robust return predictor that appears to contain information orthogonal to a large range of other predictive variables