I developed an automated statistical trading system that pulls live price data through APIs and uses historical market data to score signals, optimize portfolios, and support multi-venue execution. The system uses intraday indicators to respond to movements within the trading day.
To address drawdown and improve how the system captures price excursions, I added MFE decay exit logic and rolling peak tracking, and tuned buy thresholds and exit parameters. I also worked on adaptive signal confidence and exit-lag learning, using logged trade data and regime metadata to inform trading heuristics.
I’m developing rolling statistical calibration for price features and testing whether the spread between fast and historical probability signals contains predictive information. My related interests include quantum computing in finance, including exploring probability learners as quantum circuits with IBM Qiskit.
