New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add DataHedger and DeepHedger #575
Conversation
justinhou95
commented
Apr 13, 2022
- Use a robust expression of entropic_risk_measure for large values.
- Add Data hedger using any generated data and features.
- Add Deep hedger using different neural network structures in different time steps.
- Add a notebook illustrating the use of data hedger and deep hedger
- Add matplotlib into dependencies
- Modify the entropic risk measure to be more robust for large price values - Build datahedger class which can hedge with data already simulated or generated - Build deephedger class which can hedge with different neural networks at different time steps - add matplotlib into dependency (maybe move to dev-dependencies next time we do not need to include plotting features in the future) - add data_hedge.ipynb illustrating the two new classes
use only entropic risk measure
Oh... I found almost the same modification of entropic_risk_measure. |
BTW, although I have not fully checked your data hedger, at glance, it seems to be not generalized enough. |
Hi! Thank you so much for your reply! I wrote this data hedger mainly due to the following concerns:
|
I can understand your motivation for this implementation, but from the perspective of the design of pfhedge, it still seems to be weird, at least for me. Thus, even if we implement a kind of data hedger, we should respect and follow the design of the original pfhedge. First, if we want to use the actual price movement, we can implement like
If you want to use the generator of GAN,
Those are roughly written; thus, I don't guarantee those are working. Using the same procedure, the problem of "Calculation of the Feature" is also solved. Of course, I agree with that the current implementation has a great burden on using those data. But, I think we should implement following the pfhedge architecture for them as a part of this package. Lastly, about the last problem (flexibility in NN), I think the current pfhedge has enough flexibility. I have also noticed the problem that the hedging model should depend on the time-to-maturity, which might motivate you to implement the switching model architecture depending on timestep. Although I wrote a lot, it doesn't intend to deny your motivation and your implementation. |
Thank you so much for your detailed reply! Now it makes great sense to me why we are persisting within the structure of current |