This is my PyTorch implementation for the paper:
Wang, S. , & Hong, L. J. . (2021). Option pricing by stochastic differential equations: A simulation optimization approach.
Here is the example of 50ETFcall option dataset.
Best Iter=[10]@[172.9] test_MAE=[0.011914609879762134]
Hope it can help you.
The code has been tested under Python 3.6.7. The required packages are as follows:
- pytorch == 1.10.2
- numpy == 1.19.5
- scipy == 1.1.0
- pandas == 1.0.5
- pyDOE == 0.3.8
The instruction of commands has been clearly stated in the codes (see the parser function in Utility/parser.py).
python main.py --lr 1e-3 --save_flag 1 --epoch 400 --train_rate 0.8