We present a nonparametric method to recover a bound on ex ante dispersion of beliefs (DBB) from asset prices with minimal assumptions. DBB constrains the dispersion among all possible distributions in an economy, consistent with observed prices and subject to a good-deal bound. In model-based economies, DBB effectively tracks belief heterogeneity and serves as a diagnostic tool for evaluating model calibrations. Empirically, DBB relates to common proxies of belief dispersion, offering a real-time, market-implied disagreement measure. Our versatile approach applies to both complete and incomplete markets represented by any asset class.
This paper was accepted by Kay Giesecke, finance.
Supplemental Material:
The online appendix and data files are available at
https://doi.org/10.1287/mnsc.2022.01587
....

Dispersion of Beliefs Bounds: Sentimental Recovery

10.1287/mnsc.2022.01587