gavindiaz/awesome-quant
1
0
Fork 0
mirror of https://github.com/wilsonfreitas/awesome-quant.git synced 2026-07-27 20:07:50 +00:00
Code Issues Packages Projects Releases Wiki Activity

Add purgedcv to Trading & Backtesting (#418)

Browse Source
This commit is contained in:
Evgenii Lazarev
2026-07-20 13:56:45 +03:00
committed by GitHub
parent 9caff9aac8
commit b97ba74f6d
1 changed files with 1 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+1
View File
@@ -160,6 +160,7 @@ A curated list of insanely awesome libraries, packages and resources for Quants
- [Wickra](https://github.com/wickra-lib/wickra) - `Rust` `Python` `JavaScript` `C++` `C#` `Golang` `Java` `R` - Streaming-first technical-analysis library with a Rust core: 514 indicators updating in O(1) per tick, with bit-exact batch-vs-streaming results.
## Trading & Backtesting
- [purgedcv](https://github.com/eslazarev/purged-cross-validation) - `Python` - scikit-learn-compatible purged, group-purged, and combinatorial purged (CPCV) cross-validation, walk-forward splitting, and backtest-overfitting statistics (deflated and probabilistic Sharpe ratios, PBO, minimum backtest length) to prevent leakage and overfitting when backtesting trading strategies.
- [AlgoVault](https://github.com/AlgoVaultLabs/crypto-quant-signal-mcp) - `TypeScript` - MCP server returning composite crypto trade verdicts (direction, confidence, regime) across 5 perpetual-futures venues, with cross-venue funding-rate arbitrage and an on-chain Merkle-verified track record. Free tier.
- [capitalcom-cli](https://github.com/SimonTarara62/capitalcom-cli) - `Python` - Unofficial CLI and async SDK for the Capital.com broker API: market data, guarded order execution, and real-time streaming.
- [Inalpha](https://github.com/mirror29/inalpha) - `Python` `TypeScript` - Conversational multi-agent quant framework where agents rank currently-effective factors for entry timing (time-series rank IC), write complete strategy code that passes sandboxed audit before backtesting, and evolve strategies under multi-objective fitness; every order requires machine approval and the LLM never has a direct order path.