Ambiguous Text

Abstract

Investors infer ambiguity from text in news and social media. A proxy for information
ambiguity is developed from text processing and used in regression tests against
the S&P 500 returns. A risk neutral agent model with uniform prior beliefs is developed
to explain the ambiguity premium/discount under unfavourable/favourable
market conditions agnostic of his ambiguity preferences. The model postulates the
ambiguity premium is often elusive in efficient markets due to returns unpredictability,
and the information ambiguity as an omitted variable bias in the fundamental
relationship between risks and returns. Empirically, the paper finds that the news
media drive equity prices more than the social media except from Jun 2009 to Nov
2016..

Keywords: Social media, mixture of distributions, natural language processing, ambiguity aversion, reflection, fundamental relation

JEL Classification: D10, D12, G40, G41