Applied Mathematics and Nonlinear Sciences
Journal license

Journal

Applied Mathematics and Nonlinear Sciences


Volume
& Issue

Volume 8, Issue 2


Published
on

July 24, 2023


Pages


DOI

Article

Simulation analysis of asset pricing based on the Gaussian process

Check for updates


Authors

Jingping Nong Affiliation:
School of Reconnaissance, Guangxi Police College, Nanning, Guangxi, 530022, China.


Abstract

This paper simulates and analyzes asset pricing based on the Gaussian model, simulates stock price paths under the Heston model, B-S model, and Gaussian model using the Monte Carlo simulation method, and compares with the real path. The difference between the Heston model, the B-S model, and the real value for predicting KIA financial assets is greater than 0.6, while the difference between the Gaussian model and the real value is less than 0.25. The difference between the Heston model, the B-S model, and the true value of Shenzhen Holdings’ financial assets is greater than 1, while the difference between the Gaussian model and the true value is less than 0.35. Similarly, the difference between the Heston model, the B-S model, and the true value of Shun Tak Group’s financial assets is almost greater than 0.60, while the difference between the Gaussian model and the true value is less than 0.35. It can be seen that the Gaussian model is closer to the true value than the Heston model and the B-S model, so it is reasonable and effective to use the Gaussian model to price financial assets. This study not only presents a new basis for theoretically studying financial asset pricing but also provides a theoretical basis for applying option pricing in risk management.


Keywords

Gaussian model, Asset pricing, Monte Carlo simulation method, Heston model, B-S model, 68T01


Citation

Nong, J. (2023). Simulation analysis of asset pricing based on the gaussian process. Applied Mathematics and Nonlinear Sciences, 8(2). https://doi.org/10.2478/amns.2023.2.00073

Published by: Engineering Journals

Engineering Journals Logo