Github for Bachelor Thesis of Vinzenz Lehner May 2021
This thesis develops an extension of the Heston-Nandi GARCH model. For empirical applications it specifies two different return distribution as well as two different pricing kernels. This results in three different pricing models, which are all calibrated on European call options written on S&P 500 indexes. Additionally, a small deep learning exercise is performed to investigate the speed up in pricing function evaluations. The results of this deep learning exercise are unfortunately not reliable due to computational constraints.