We develop an innovative deep neural network (DNN) supervised learning approach to extracting insightful topic sentiments from analyst reports at the sentence level and incorporating this qualitative knowledge in asset pricing and portfolio construction. The topic sentiment analysis is performed...

We test the hypothesis that low visibility shocks to text-based network industry peers can explain industry momentum. We consider industry peer firms identified through 10-K product text and focus on economic peer links that do not share common SIC codes. Shocks to less visible peers generate...