"""
Quantitative Financial Models for Cryptocurrency Markets.

This module implements sophisticated financial models adapted for crypto:
- GARCH family volatility forecasting
- VaR/CVaR/ES risk measures
- Hamilton filter regime detection
- Jump detection (Lee-Mykland)
- Kalman filter for dynamic hedge ratios
- Black-Scholes options pricing
- EVT for tail risk
- Market microstructure models

All models follow the LeadLagBrain pattern:
- O(1) on_tick() updates
- EMA-based statistics
- Dashboard integration via get_stats()
"""

from .base import (
    BaseModel,
    MultiSymbolModel,
    ModelSignal,
    SignalType,
    SignalCallback,
)

# Volatility models
from .volatility.garch import GARCH11, EGARCH
from .volatility.realized_vol import RealizedVolatility

# Risk models
from .risk.var_es import RealTimeVaR, EVTVaR

# Regime detection
from .regime.hamilton_filter import HamiltonFilter

# Jump detection
from .jumps.jump_detection import LeeMyklandTest

# Statistical arbitrage
from .stat_arb.kalman_filter import KalmanHedgeRatio

# Options pricing
from .options.black_scholes import BSMPricer, BSMGreeks

# Microstructure
from .microstructure.price_impact import KyleLambda

# Model manager
from .manager import ModelManager, ModelConfig

__all__ = [
    # Base
    "BaseModel",
    "MultiSymbolModel",
    "ModelSignal",
    "SignalType",
    "SignalCallback",
    # Volatility
    "GARCH11",
    "EGARCH",
    "RealizedVolatility",
    # Risk
    "RealTimeVaR",
    "EVTVaR",
    # Regime
    "HamiltonFilter",
    # Jumps
    "LeeMyklandTest",
    # Stat Arb
    "KalmanHedgeRatio",
    # Options
    "BSMPricer",
    "BSMGreeks",
    # Microstructure
    "KyleLambda",
    # Manager
    "ModelManager",
    "ModelConfig",
]