A Binomial Tree-Based Empirical Study on the Biases of American Call Option

Abstract

The application of financial derivatives, particularly options, has become pervasive in modern finance, offering effective tools for risk management. Among these, American options, allowing flexibility in exercising rights prior to expiration, dominate the derivatives market. Accurate pricing of American options is crucial for informed investment decisions and risk assessments. While various pricing models exist, the binomial tree model is welcomed for its simplicity and accessibility. However, inherent biases in pricing models raise questions about their efficacy across different sectors of the market. Based on the binomial tree model, this study empirically examines the pricing accuracy of American call options in both the technology sector and the industrial sector, by collecting data from top five largest companies in each sector. For the technology sector, samples are mainly from semiconductor manufacturers, software companies, etc., which are all situated in thriving fields; while those five companies from the industrial sector mainly belong to traditional industries, such as energy and engine suppliers. By analyzing prediction biases between two sectors, this research tries to give out reasons behind it and grasps a deeper understanding of option pricing nuances between sectors, and also the limitations of the binomial tree model.