heston model calibration: implementation and empirical evaluation

This paper offers a detailed and pedagogical examination of the heston stochastic volatility model, a framework that continues to influence both theoretical research and practical applications in option pricing. by modeling volatility as a stochastic process, the heston formulation addresses empirical features of financial markets that the classical black-scholes model cannot explain, including volatility clustering, skewness, and heavy-tailed return distributions. the discussion is organized around three core components: the model’s theoretical foundation, its computational implementation, and its empirical calibration to spy option data. particular attention is given to the interaction of model parameters and the numerical challenges involved in estimation. analytical pricing is introduced through the fourier-based semi-closed form solution, while the calibration is carried out using monte carlo simulation combined with a gradient descent optimization scheme. the paper ultimately seeks to bridge the conceptual structure of the model with its practical interpretation, emphasizing how stochastic volatility theory connects mathematical modeling to observable market behavior.