Ledger is a command-line toolkit that evaluates retail trading strategies using walk-forward validation rather than a single train/test split. It models per-instrument execution costs, audits look-ahead leakage, applies purge windows, and simulates tier-based exits for quantitative developers before live trading.
Ledger is a Trading bots, indicators & backtesting project. It focuses on avoiding misleading backtests and unexpected losses when deploying algorithmic trading strategies. Ledger is a B2B product aimed at retail algorithmic traders and quantitative developers. Ledger is available on the command line.
Key capabilities include walk-forward validation, rolling test windows, and cost modeling.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Same category — not a similarity match