Skip to content

Latest commit

 

History

History
79 lines (58 loc) · 3.45 KB

TODO.adoc

File metadata and controls

79 lines (58 loc) · 3.45 KB
Raw

Release planning

Introduction

There are still many features planned for roboquant. For sure, the lists on this page are also not complete and new features will pop up over time. Prioritization might shift based also based on community input and contribution.

But at least it gives insight into the current planning when to add certain features. And PRs are very welcome and might very well expedite certain features.

See also the contributing page on how to submit a PR.

Version 0.8.x (end of 2021)

Version 0.8 is all about making sure that (back-)testing works, and the most common use-cases are covered. Much of the foundation for algo trading is in place with this release:

  • ✓ Improve the documentation in source code

  • ✓ Improve unit test coverage

  • ✓ Add visualizations for the Jupyter notebooks

  • ✓ Improve CSV parsing with extra configurable options

  • ✓ Improve error messages and warnings to be more helpful

  • ✓ Add documentation on how to install and get started

  • ✓ Support for advanced order types

  • ✓ Add info on key design concepts behind the software, so it becomes easier to contribute

  • ✓ Bring back Interactive Brokers integration

  • ✓ See how to best fit crypto trading with current account structure

Version 0.9.x (2022)

Ensure the API is stable and start including AI and Machine Learning support

  • [X] Improve documentation, notebooks, and examples

  • [X] Include demo feeds for quick experimentation

  • [X] Use better approach for generating charts

Version 1.0 (end of 2022)

This version is all about adding stabilizing concepts and APIs. Although there is already integration available in earlier versions, that is just to validate the architectural concepts and design choices:

  • [X] Improve the test coverage

  • [X] Stabilize the exposed APIs

  • [X] Optimize performance and avoid regressions

  • [X] Refactor the order execution simulation

  • [X] Improve the running of roboquant notebooks on public infra

Version 2 (Q3 2023)

The topics mentioned here are some of the ideas for the future releases:

  • [X] Add common back test approaches

  • [X] Improve performance for specific use cases

  • [X] Improve the default FlexPolicy

  • [X] More flexible charting

  • [X] Support for more feed formats (QuestDB)

  • [X] Easier running of code on remote machines (web server, docker)

  • [X] Separate charts functionality from Jupyter notebooks

  • [X] Improve the Broker API

  • [X] Smaller core module

  • [X] More consistent naming of API’s

  • [X] Support more CSV file formats out of the box

  • [X] Improved usability of TaLib with auto sizing

  • [X] Sunset no longer suported 3rd party data providers

  • [X] Optimize memory usage for large back tests

  • [X] Improve overall test coverage

  • [X] Hardening code base

Version 3 and later (2024 and beyond)

The topics mentioned here are some of the ideas for the future releases:

  • ❏ Support Kotlin 2.0

  • ❏ Add support for reading different types of data feeds besides price actions

  • ❏ Better and more complete integrations with brokers and crypto-exchanges

  • ❏ Advanced policies, like auto re-balancing portfolios

  • ❏ Machine-learning-based strategies

  • ❏ Right now messages support the English format only, add L10N/I18N support

  • ❏ Make video(s) that show the steps to develop and test your own strategies

  • ❏ Come up with a way users can easily share strategies, code snippets and other best practices

  • ❏ Better support for displaying multiple metrics/indicators on a single chart