google / google/tf-quant-finance
JointItoProcess
- Dominant language
- Python
- Stars
- 5.5k
- Forks
- 698
- PR merge metrics
- No merged PRs in 30d
Description
Hello,
it seems that the correlation structure of the JointItoProcess model cannot accept a different structure from that of block diagonal across the single processes. I am dealing with joint Heston Processes and ideally I would like to introduce correlation within single spot price processes as well as correlation within single variance processes (and also correlation within variances and spot prices of different process). That is not possible with current release correct (I am just trying to understand is not really request)? Thanks
Contributor guide
Research direction
The issue names JointItoProcess and joint Heston processes but does not identify a file or test. Start by locating JointItoProcess and reading how its correlation structure is restricted to block-diagonal single processes. Done would mean establishing and implementing a clear design for the requested within-process and cross-process correlations, with corresponding coverage.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, tensorflow
- Domain
- fintech-quant
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100