Option Pricing Based on the Residual Neural Network

Gan，Lirong1; Liu，Weihan2

2023

10.1007/s10614-023-10413-3

Computational Economics   影响因子和分区

0927-7099

1572-9974

We employ an innovative deep learning method to price options quickly and accurately. Specifically, we construct the Residual Neural Network model (ResNet) by two different basic residual blocks with three one-dimensional convolution layers and a shortcut. This model is a generalized option pricing method, and it can be used to approximate the option pricing formula without any assumptions. Besides, the model also can be easily extended to the deep ResNet model to achieve higher prediction accuracy. Comprehensive numerical experiments show that the deep ResNet model has excellent performance in the pricing of 50ETF options in the Chinese market, and the prediction accuracy of our model is higher than that of commonly used deep learning models, including deep neural network (DNN) and fully convolutional networks (FCN).

Deep network Machine learning Option pricing ResNet

SCI ; SSCI

英语

其他

National Natural Science Foundation of China[72031003] ; Humanities and Social Sciences Project of the Ministry of Education of China[22YJA790067] ; Cultivation of Guangdong College Students' Scientific and Technological Innovation[pdjh2023c31901]

Business & Economics ; Mathematics

Economics ; Management ; Mathematics, Interdisciplinary Applications

WOS:001042470100002

SPRINGER

2-s2.0-85166616401

Scopus

1.School of Financial Mathematics & Statistics,Guangdong University of Finance,Guangzhou,No. 527, Yingfu Road, Tianhe District,510521,China
2.Department of Finance,Southern University of Science and Technology,Shenzhen,1088 Xueyuan Avenue,518055,China

Gan，Lirong,Liu，Weihan. Option Pricing Based on the Residual Neural Network[J]. Computational Economics,2023.

Gan，Lirong,&Liu，Weihan.(2023).Option Pricing Based on the Residual Neural Network.Computational Economics.

Gan，Lirong,et al."Option Pricing Based on the Residual Neural Network".Computational Economics (2023).