Multi-asset pricing modeling using holding-based networks in energy markets

This paper studies the multi-asset pricing problem in the stock market with holding-based networks. We derive a linear noisy rational expectation equilibrium (LNREE) in the network, construct holding-based networks and test the LNREE's efficiency in generating predictive output validation. The results show that the LNREE fits the observed prices and returns of stocks quite well, especially for U.S. energy stocks. By changing the average risk aversion, the predictive performance of the model can be optimized. It provides a novel perspective for the empirical study on multi-asset pricing modeling.