Based on a family of discrepancy functions, we derive nonparametric stochastic discount factor bounds that naturally generalize variance, entropy, and higher-moment bounds. These bounds are especially useful to identify how parameters a ect pricing kernel dispersion in asset pricing models. In particular, they allow us to distinguish between models where dispersion comes mainly from skewness from models where kurtosis is the primary source of dispersion. We analyze the admissibility of disaster, disappointment aversion, and long-run risk models with respect to these bounds.
All Science Journal Classification (ASJC) codes
- Strategy and Management
- Management Science and Operations Research
- Implicit utility maximizing weights
- Information-theoretic bounds
- Minimum contrast estimators
- Stochastic discount factors