Traditional sentiment construction in finance relies heavily on the dictionary-based approach, with a few exceptions using simple machine learning techniques such as Naive Bayes classifier. While the current literature has not yet invoked the rapid advancement in the natural language processing, we construct in this research a textual-based sentiment index using a novel model BERT recently developed by Google, especially for three actively trading individual stocks in Hong Kong market with hot discussion on Weibo.com. On the one hand, we demonstrate a significant enhancement of applying BERT in sentiment analysis when compared with existing models. On the other hand, by combining with the other two existing methods commonly used on building the sentiment index in the financial literature, i.e., option-implied and market-implied approaches, we propose a more general and comprehensive framework for financial sentiment analysis, and further provide convincing outcomes for the predictability of individual stock return for the above three stocks using LSTM (with a feature of a nonlinear mapping), in contrast to the dominating econometric methods in sentiment influence analysis that are all of a nature of linear regression.
↧