[Research] (Re)define research questions

[mjuchli]

Research questions have to be redefined precisely. Let's try to define the questions such that the experiments done so far contribute towards answering the questions substantially.

Furthermore, a high-level process pipeline should be laid out which not only serves as an overview of the capabilities of the final product but also leads me to work targeted towards answering the research questions and build a fully functional executioner.

[mjuchli]

A draft has been finalized and is ready to be discussed at the meeting:

• RQ: https://github.com/backender/ctc-executioner/wiki/Home/1306330d48d65676f9edf51053ee8cca8f007878
• Pipeline: https://github.com/backender/ctc-executioner/wiki/5.-Optimization/0f17876cce6b98b019842a4337bbf944026e0a72

[mjuchli]

The main research questions has been redefined as:

How can data, which is derived from a limit order book, be used as features in a reinforcement learning environment in order to contribute to the optimization of order execution?