Multi-Asset Pricing Modeling Using Holding-Based Network in Energy Markets
Wentao Wang,Junhuan Zhang
DOI: https://doi.org/10.2139/ssrn.3728499
2019-01-01
SSRN Electronic Journal
Abstract:This paper combines multi-asset pricing model with network theory to study multiasset pricing in the holding-based network. We obtain a new expression of equilibrium price by inducing the network parameter. To testify the practical significance of our model of real asset prices, we fit the quarterly prices of stocks from four energy indices and measure the fitting effect. Firstly, we analyze the evolution of network properties and find: the numbers of shareholders of the CSI 300 Energy stocks and the S&P 500 Energy stocks are more stable than those of whom invest in the SSE Energy Sector stocks and the SZSE Energy Sector stocks; in addition, the networks from the more stable stocks are denser than those from the less stable one; the market crash during the period of June to September 2015 causes a critical drop in the quantity of shareholders and increases in shareholders’ co-holding behavior and stability; the shareholders of American listed energy companies have strengthened co-holding behavior and higher stability than those of China’s listed energy companies. Secondly, by defining the fitting coefficient F evaluating the fitting effect under different risk aversion scenarios, we argue that the investors of the SSE Energy Sector stocks, the SZSE Energy Sector stocks, the CSI 300 Energy stocks and the S&P 500 Energy stocks, are respectively risk-averse, risk-neutral, risk-loving and risk-neutral. Thirdly, by plotting the PDFs and CCDFs of stock returns, compared to the Gaussian distribution, we find that both the fitted returns and the real returns depict the features of high kurtosis and fat tail. Eventually, the comparison shows that, in diverse levels of risk aversion for each energy index, the approach proposed in this research is more accurate in fitting relative to the conventional method.