Bayesian Inference and Correntropy for Asset Pricing Under Ambiguity

Summary

The paper proposes a Bayesian Inference and Correntropy technique for modeling uncertainty and ambiguity in market information and evaluating asset prices. It applies the approach to a consumption-based asset-pricing model, representing consumption variation with a Bayesian network and examining price dynamics in data.

The reported dynamics include procyclical price deviations, countercyclical equity premia and volatility, leverage effects, and mean reversion in excess returns. The authors state that modeling asset information with this technique can estimate market price changes effectively. The supplied description does not specify the data, benchmark methods, estimation procedure, or quantitative evaluation, so the claimed accuracy and generality cannot be assessed from this account alone.

Key ideas

• The proposed BIC technique combines Bayesian inference and correntropy to represent uncertainty and ambiguity.
• A consumption asset-pricing model uses a Bayesian network to represent consumption variation.
• The analysis considers price deviations, equity premia, volatility, leverage effects, and excess-return mean reversion.
• The paper reports effective estimation of price changes but provides no metrics or benchmark comparisons in the supplied description.

Tags

# Accurate Evaluation of Asset Pricing Under Uncertainty and Ambiguity of Information


# Accurate Evaluation of Asset Pricing Under Uncertainty and Ambiguity of Information









Since exchange economy considerably varies in the market assets, asset prices have become an attractive research area for investigating and modeling ambiguous and uncertain information in today markets. This paper proposes a new generative uncertainty mechanism based on the Bayesian Inference and Correntropy (BIC) technique for accurately evaluating asset pricing in markets. This technique examines the potential processes of risk, ambiguity, and variations of market information in a controllable manner. We apply the new BIC technique to a consumption asset-pricing model in which the consumption variations are modeled using the Bayesian network model with observing the dynamics of asset pricing phenomena in the data. These dynamics include the procyclical deviations of price, the countercyclical deviations of equity premia and equity volatility, the leverage impact and the mean reversion of excess returns. The key findings reveal that the precise modeling of asset information can estimate price changes in the market effectively.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.