Abstract
Based on GARCH models, the purpose of this paper is to extend the two-factor volatility components model by taking a U-shape pricing kernel which was developed by Christoffersen et al. (2013). By imposing the U-shaped pricing kernel, we surprisingly obtain a more elaborate model which can also explain some puzzles in the market. Furthermore, this article derives a closed-form solution for option valuation which is convenient and propitious for computational financing. As for the methodology used in estimation, we chose the EM algorithm to modified the likelihood function because of the incomplete data. The empirical result demonstrates the well ability of the generalized model when reconcile time series properties of stock returns with the option prices. Moreover, we also use the in-sample and out-of-sample to test the predictability of the generalized model. Under our specification, the U-shaped pricing kernel makes the new model somehow more predictable than before.