Module interbar 

Inter-bar microstructure features computed from lookback trade windows

GitHub Issue: https://github.com/terrylica/opendeviationbar-py/issues/59

This module provides features computed from trades that occurred BEFORE each bar opened, enabling enrichment of larger open deviation bars (e.g., 1000 dbps) with finer-grained microstructure signals without lookahead bias.

Temporal Integrity

All features are computed from trades with timestamps strictly BEFORE the current bar’s open_time. This ensures no lookahead bias in ML applications.

Feature Tiers

• Tier 1: Core features (7) - low complexity, high value
• Tier 2: Statistical features (5) - medium complexity
• Tier 3: Advanced features (4) - higher complexity, from trading-fitness patterns

Academic References

Feature	Reference
OFI	Chordia et al. (2002) - Order imbalance
Kyle’s Lambda	Kyle (1985) - Continuous auctions and insider trading
Burstiness	Goh & Barabási (2008) - Burstiness and memory in complex systems
Kaufman ER	Kaufman (1995) - Smarter Trading
Garman-Klass	Garman & Klass (1980) - On the Estimation of Security Price Volatilities
Hurst (DFA)	Peng et al. (1994) - Mosaic organization of DNA nucleotides
Permutation Entropy	Bandt & Pompe (2002) - Permutation Entropy: A Natural Complexity Measure

Structs

InterBarConfig

Configuration for inter-bar feature computation

InterBarFeatures

Inter-bar features computed from lookback window

TradeHistory

Trade history ring buffer for inter-bar feature computation

TradeSnapshot

Lightweight snapshot of trade for history buffer

Enums

LookbackMode

Lookback mode for trade history