Abstract
This paper studies option pricing based on a reverse engineering (RE) approach. We utilize artificial intelligence in order to numerically compute the prices of options. The data consist of more than 5000 call- and put-options from the German stock market. First, we find that option pricing under reverse engineering obtains a smaller root mean square error to market prices. Second, we show that the reverse engineering model is reliant on training data. In general, the novel idea of reverse engineering is a rewarding direction for future research. It circumvents the limitations of finance theory, among others strong assumptions and numerical approximations under the Black–Scholes model.
Highlights
The most remarkable scientific breakthrough in recent years is the creation of super-human intelligence via deep neural networks in the field of artificial intelligence (AI) (Silver et al 2016, 2017, 2018; Tian et al 2019)
This paper studies option pricing based on a reverse engineering (RE) approach
This paper focuses on a certain financial derivative, namely options
Summary
The most remarkable scientific breakthrough in recent years is the creation of super-human intelligence via deep neural networks in the field of artificial intelligence (AI) (Silver et al 2016, 2017, 2018; Tian et al 2019). Given the enormous scale and complexity of data, a manual human analysis is impossible. We utilize the tools of AI and apply them to finance, on pricing financial derivatives. This paper focuses on a certain financial derivative, namely options. With the aim to simplify the subsequent analysis, we first study plain vanilla call- and put-options. A call(put)-option gives the holder the right to buy (sell) an underlying asset at a specified strike price and time in the future. European options can only be exercised at the end of maturity, while American-type options can be exercised at any time
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have