Forecasting Call Option prices : A Quantitative Study in Financial Economics

Sammanfattning: It is not uncommon that the theoretical price of a model is different from the market price due to various disturbances. The purpose of this study was to analyze how well the original Black-Scholes-Merton model performs accurate forecasts of the option price, where the underlying asset was the NIFTY50 stock index. The essential idea was to compare a more original complex model with time series models, where the series only depends on the option price itself. The analysis ended up with comparing the Black-Scholes-Merton model to the AR-GARCH and two different orders of ARMA-GARCH model. The result indicates that the Black-Scholes-Merton model was the most accurate and best-performed in 2018. On the contrary, during 2019, the Black-Scholes-Merton model was the worst-ranked in performance of accuracy. Because the model fit of the forecasted variables in the Black-Scholes-Merton model was built based on historical train data in late 2017, this is most likely the reason behind the more mediocre performance during 2019. Overall the AR-GARCH became the highest performed model in forecast accuracy.