Most existing text-based sentiment measures in finance are lexicon-based which are effectively based on word counts of positive and negative sentiment dictionaries, and naturally lose most information. We measure news sentiment using BERT, a state-of-the-art large language model, which reads and...

Results from a large sample of individual Chinese investors demonstrate that they were more likely to trade stocks for short-term speculation after experiencing trauma such as natural disasters, serious illness, or death in their immediate family. They exhibited higher impulsivity, a greater...

We use brokerage account records to study trading during the Chinese put warrants bubble and find evidence consistent with extrapolative theories of speculative asset price bubbles. We identify the event that started the bubble and show that investors engaged in a form of feedback trading based...