How ca-manmohan/heston-0930-option-buyer works
This repository describes a NIFTY 50 weekly ATM-options strategy that evaluates causal 09:30 telemetry, chooses a CE/PE trade or stands aside, and manages position sizing and exits through 11:30. The research script builds historical features and the backtest evaluates symbolic rules and simulated trades. The README also documents a GPU actor–critic discovery workflow, but no implementation for it appears in the supplied source tree, so it is represented only as a documented capability. Backtest data wiring beyond the shown code is not inferred.
Main entry points: strategy_eas_heston_0930.py Trade gate · research_heston_stress_persistence_strategy.py Session features · backtest_heston_stress_persistence_20k.py Rule evaluation · symbolic_heston_0930_invariant.py Symbolic invariant
Components 10
Research
- Session features
research_heston_stress_persistence_strategy.py - Heston calibration
research_heston_stress_persistence_strategy.py
Signal and Execution
- Symbolic invariant
symbolic_heston_0930_invariant.py - Trade gate
strategy_eas_heston_0930.py - Position sizing
strategy_eas_heston_0930.py - Tranche exits
strategy_eas_heston_0930.py
Backtest
- Rule evaluation
backtest_heston_stress_persistence_20k.py - Capital backtest
backtest_heston_stress_persistence_20k.py - GPU actor–critic
Other components
- NSE telemetry
Connections 7
- NSE telemetry → Session features: supplies telemetry
README.md - Session features → Heston calibration: calibrates slices
research_heston_stress_persistence_strategy.py - Trade gate → Symbolic invariant: evaluates signal
strategy_eas_heston_0930.py - Trade gate → Position sizing: sizes position
strategy_eas_heston_0930.py - Trade gate → Tranche exits: simulates exits
strategy_eas_heston_0930.py - Rule evaluation → Capital backtest: scores candidates
backtest_heston_stress_persistence_20k.py - GPU actor–critic → Rule evaluation: discovers rules
README.md