A1063
Title: Economy-driven consumption-based asset pricing model
Authors: Anastasija Tetereva - Erasmus University Rotterdam (Netherlands) [presenting]
Alberto Quaini - Erasmus University of Rotterdam (Netherlands)
Abstract: The standard consumption-based asset pricing model, while theoretically significant, often fails to explain the cross-sectional patterns of returns observed in practical datasets. The limitation is frequently attributed to missing factors, especially in connection with the omission of conditioning information. That is, merely relying on consumption might not be sufficient to capture the factor structure underlying asset returns across all economic states. To address the issue, an extension to the traditional model is proposed by leveraging the machine learning technique of regression trees. The approach aims to develop a local market condition-driven consumption-based asset pricing model. By integrating additional factors in key local market conditions, the model significantly enhances the ability to explain the cross-section of stock returns in the subsequent period. By considering specific local factors, the unique market dynamics are captured that contribute to a comprehensive understanding of the underlying forces influencing asset pricing. In conclusion, expansion of the scope of the consumption-based asset pricing model is achieved, and valuable insights into the dynamic nature of markets are provided, paving the way for improved investment decision-making and risk management strategies.
