search

microsoft / qlib

Qlib is an AI-oriented quantitative investment platform that aims to realize the potential, empower research, and create value using AI technologies in quantitative investment, from exploring ideas to implementing productions. Qlib supports diverse machine learning modeling paradigms. including supervised learning, market dynamics modeling, and RL.
https://qlib.readthedocs.io/en/latest/
MIT License
14.55k stars 2.53k forks source link

Futures Trading #107

Open ghost opened 3 years ago

ghost commented 3 years ago

🌟 Feature Description

Futures trading support

Do you have any plans to add the futures trading support to the framework?

you-n-g commented 3 years ago

Yes. Qlib is designed to support all financial derivatives include futures. But futures trading is quite a broad and somewhat fuzzy word. Can you give some examples of your future trading research works and what features are required to support your research?

nkiki commented 3 years ago

Yes. Qlib is designed to support all financial derivatives include futures. But futures trading is quite a broad and somewhat fuzzy word. Can you give some examples of your future trading research works and what features are required to support your research?

how about cryptocurrency?:)

you-n-g commented 3 years ago

@nkiki I think most of the research works on cryptocurrency are similar to alpha forecasting and High-frequency trading . For these two cases, the former has been alreadly supported by Qlib's framework areadly (But the data is not included.) The latter will be supported in the future.

ghost commented 3 years ago

Yes. Qlib is designed to support all financial derivatives include futures. But futures trading is quite a broad and somewhat fuzzy word. Can you give some examples of your future trading research works and what features are required to support your research?

Like reinforcement learning based "Medium-frequency" trading strategies

you-n-g commented 3 years ago

@wycfutures The RL diagram is not supported in the current version, but it is what we are designing and we will try to implement a framework in the future.
Welcome to discuss with us about the design of framework.

ghost commented 3 years ago

I'd be happy to see the existing framework support the S&P 500 E mini: https://finance.yahoo.com/quote/ES=F/

ChengYen-Tang commented 3 years ago

@nkiki I think most of the research works on cryptocurrency are similar to alpha forecasting and High-frequency trading . For these two cases, the former has been alreadly supported by Qlib's framework areadly (But the data is not included.) The latter will be supported in the future.

Hi @you-n-g , Are there any plans to add the cryptocurrency trading environment in the backtest? For example: Okex Perpetual Swap ?

you-n-g commented 3 years ago

@ChengYen-Tang Currently, we are working on developing high-frequency trading based on supervised learning. A framework and some simple examples will be included. You can check if your requriements can be fulfilled.

ChengYen-Tang commented 3 years ago

Hi @you-n-g , I have a question. Does the backtest include short-term operations? Also, are there relevant documents that include the backtest environment?

you-n-g commented 3 years ago

Hi, @ChengYen-Tang Here are some documents about backtest environment (some interfaces may be refactored recently). https://qlib.readthedocs.io/en/latest/component/backtest.html

Short-term operation may be supported in the next version

ChengYen-Tang commented 3 years ago

OK, Hope this problem can be resolved as soon as possible, thank you. https://github.com/microsoft/qlib/issues/237