Title: A tale of sentiment driven tail events: A dynamic quantile model for asset pricing with sentiment
Authors: Jozef Barunik - UTIA AV CR vvi (Czech Republic) [presenting]
Cathy Yi-Hsuan Chen - University of Glasgow (United Kingdom)
Wolfgang Karl Haerdle - Humboldt University at Berlin (Germany)
Abstract: The link between investor sentiment and asset valuation is the subject of considerable debate in the profession. The aim is to abandon the classical asset pricing that relies on expected utility, and introduce a dynamic quantile model for asset pricing, in which the agent maximizes stream of the future quantile utilities instead. Using the model, we empirically investigate if investor sentiment distilled from textual mining analysis can price tails of the return distributions. On the panel of 100 stocks, we document influence of aggregate investor's sentiment on future conditional quantiles of the return distributions. Aggregate sentiment explains cross-section of tails even after controlling for popular factors used in the literature, as well as firm-specific sentiment and volatility.