Quote Engine Integration

To generate real-time quotes, you wire model components like OnlineHMMFilter and Hawkes intensity into a deterministic pipeline. Feeding MarketData per tick into the QuoteEngine produces a QuoteSet with final quotes and diagnostics.

End-to-end quote generation uses a deterministic 11-step pipeline that computes kappa, gamma, and final quotes. It integrates market data feeds, model blocks, and a parameter blending layer to output QuoteSets with calibration logging.

To add a new model component, integrate it into the configurable parameter blending layer within the quote generation pipeline. The architecture supports adding blocks like Adverse Selection Adjuster or RegimeConditionedLeadLag to process live or replay data.

Yes, you can debug a multi-model pipeline across simulated or replay data. The Quote Engine Integration provides logging for calibration and testing, allowing you to trace issues through model blocks during the quote generation process.

Yes, a configurable parameter blending layer is required to produce QuoteSet outputs. It works with integrated model blocks and data feeds to calculate the final quotes and diagnostics across live or simulated data.

An Adverse Selection Adjuster is needed in quote generation to manage risk by adjusting quotes based on adverse selection metrics. It acts as a model block within the pipeline to refine the final QuoteSet output.