Skip to content
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

highfreq test dataset range? #701

Closed
2young-2simple-sometimes-naive opened this issue Nov 18, 2021 · 2 comments
Closed

highfreq test dataset range? #701

2young-2simple-sometimes-naive opened this issue Nov 18, 2021 · 2 comments
Assignees
Labels
question Further information is requested

Comments

@2young-2simple-sometimes-naive
Copy link
Contributor

I have tried to run the highfreq workflow using the given example qlib_data downloaded (yahoo_cn_1min). The MARKET = "all" gives an IC of around 0.008, which is far from the reported 0.3 in the example. In fact I don't believe IC~0.3 could be achieved with ~4000 symbols. Is the example actually using csi300 market?

@2young-2simple-sometimes-naive 2young-2simple-sometimes-naive added the question Further information is requested label Nov 18, 2021
@zhupr
Copy link
Collaborator

zhupr commented Nov 20, 2021

@2young-2simple-sometimes-naive Hi,
The example of highfreq is csi300 market, which has been fixed for the results: #702
Need to re-download data: python scripts/get_data.py qlib_data --target_dir <1min dir> --interval 1min

@2young-2simple-sometimes-naive
Copy link
Contributor Author

@2young-2simple-sometimes-naive Hi, The example of highfreq is csi300 market, which has been fixed for the results: #702 Need to re-download data: python scripts/get_data.py qlib_data --target_dir <1min dir> --interval 1min

Yep. IC~0.03 is more like it.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
question Further information is requested
Projects
None yet
Development

No branches or pull requests

2 participants