In this project, calibration of parameters of Heston and Bates models using Markov Chain Monte Carlo (MCMC) is performed based on the findings in the paper by Cape et al. "Estimating Heston's and Bates’ models parameters using Markov chain Monte Carlo simulation".
There are two implementations: pure Python and cythonized Python:
1. Pure Python: the demo in demo.ipynb shows that it takes approximately 27 minutes to calibrate the Heston model with 10,000 MCMC steps and 69 minutes to calibrate the Bates model with the same number of steps.
2. Cythonized Python: the demo in demo_cythonized.ipynb shows that it takes approximately 2.4 minutes to calibrate the Heston model with 10,000 MCMC steps, that is, 11x speed-up is achieved.