Subjective valuation and target price accuracy

In this paper, we analyze how subjective adjustments to baseline models by analysts affect the forecasting accuracy. For a panel of analyst reports, we show that target price forecasts that deviate signi cantly from simple multiple-based pseudo-Target prices are (ex-post) more accurate. By controlling for various stock and broker characteristics, we also demonstrate that our results are not driven by the degree of sophistication of the valuation models. Furthermore, we show that investors know about this increased informativeness of forecasts as the abnormal market return around target price revisions is significantly higher if analysts deviate from simple pseudo-Target prices when issuing their forecasts.