goldbees-pipeline

The pipeline blends a Kelly-optimal weight with risk governance adjustments using inverse-volatility and regime-based logic. This combination produces a probabilistic signal and final position recommendation that balances return optimization with downside protection.

Yes, you can evaluate past signal calls without overwriting by using the --no-save and --evaluate flags. These flags execute the pipeline in a dry-run mode against historical data while preventing new checkpoints from being written to your database.

The pipeline blends a Kelly-optimal weight with risk governance adjustments using inverse-volatility and regime-based logic. This combination produces a probabilistic signal and final position recommendation that balances return optimization with downside protection.

The best way to apply Kelly criterion position sizing with risk governance is using a pipeline that computes the Kelly-optimal weight and blends it with inverse-volatility and regime-based adjustments. This produces a risk-governed final recommendation.

No, the GOLDBEES ML pipeline does not require external dependencies to run. It operates independently using Python scripts and built-in flags to manage signal generation, evaluation, and database checkpoint retrieval.