A Robust Bayesian Analysis of the Stock Market's Response to Macroeconomic News
46 Pages Posted: 5 Oct 2013
Date Written: October 4, 2013
This paper explores the quality of the information that macroeconomic news convey to the stock market as forward looking signals of future business conditions. We introduce a novel robust Bayesian semi-parametric analysis of investors’ correspondence functions (i.e., signal-to-price mappings) in the stock market and a feasible ex ante measure of the level of ambiguity in Survey responses anticipating macroeconomic announcements. Using both survey and vector autoregressive (VAR)-based data we show that macroeconomic announcements are relatively ambiguous signals of future economic fundamentals in the stock market, potentially explaining some of previous controversial results in the literature.
Keywords: Learning under ambiguity, Macroeconomic news, Non-parametric methods, MMS survey data
Suggested Citation: Suggested Citation