澳门威斯尼斯游戏网址

2020年11月17日学术报告通知

发布时间:2020-11-13

内容提要:

When quantifying information from unstructured textual data, the traditional bag-of-words approach  only captures semantic features of single words or phrases. The context, the sequence of words, and  the relationship between words are ignored. This paper introduces a novel approach to incorporate  complex syntactical features in the textual analysis using two deep neural network (NN)-based  methods. We construct a new measure of sentiment that is specific to performance discussions and is  adjusted for complex contextual negations. We find that this performance-specific sentiment explains  cross-sectional returns and future operating performance better than the umbrella sentiment proxies  used in the literature.