Files
xau-ai-trading-bot/src/trajectory_predictor.py
T

328 lines
12 KiB
Python
Raw Normal View History

"""
Trajectory Predictor - Prediksi pergerakan profit masa depan
Menggunakan parabolic motion model untuk forecast profit 1-5 menit ke depan
"""
import numpy as np
from typing import List, Tuple, Dict
from loguru import logger
class TrajectoryPredictor:
"""
Prediksi trajectory profit menggunakan kinematic equations.
Model: profit(t) = profit₀ + velocity*t + 0.5*acceleration*t²
Cocok untuk:
- Deteksi early exit (jangan close jika prediksi profit tinggi)
- Validasi exit timing (exit jika prediksi profit turun)
- Recovery continuation (prediksi apakah recovery akan lanjut)
"""
def __init__(self):
self.default_horizons = [60, 180, 300] # 1m, 3m, 5m (seconds)
self.confidence_threshold = 0.7 # Minimum confidence untuk pakai prediksi
# v0.2.0: Regime-based dampening factors (validated from live trades)
# Trade #161778984: avg over-prediction 7.5x -> need 85% reduction
# Trade #161850770: predicted profit from loss -> need 70% reduction
self.dampening_factors = {
"ranging": 0.20, # 80% reduction (most conservative)
"volatile": 0.30, # 70% reduction (validated: 3-17x over -> 1-5x)
"medium_volatility": 0.30, # Same as volatile
"trending": 0.50, # 50% reduction (momentum likely continues)
"normal": 0.30 # Default fallback
}
def predict_future_profit(
self,
current_profit: float,
velocity: float,
acceleration: float,
horizons: List[int] = None,
regime: str = "normal"
) -> List[float]:
"""
Prediksi profit di masa depan menggunakan parabolic motion dengan regime dampening.
Args:
current_profit: Profit saat ini ($)
velocity: Profit velocity ($/second)
acceleration: Profit acceleration ($/second²)
horizons: List of time horizons dalam seconds (default: [60, 180, 300])
regime: Market regime for dampening ("ranging"/"volatile"/"trending")
Returns:
List of predicted profits untuk setiap horizon (damped)
Example:
>>> predictor = TrajectoryPredictor()
>>> pred_1m, pred_3m, pred_5m = predictor.predict_future_profit(
... current_profit=0.05,
... velocity=0.1335,
... acceleration=0.0017,
... regime="volatile"
... )
>>> print(f"1min: ${pred_1m:.2f}, 3min: ${pred_3m:.2f}")
1min: $3.34, 3min: $8.22 (damped by 0.30x)
"""
if horizons is None:
horizons = self.default_horizons
# v0.2.0: Get dampening factor based on regime
dampening = self.dampening_factors.get(regime, 0.30)
predictions = []
for dt in horizons:
# Kinematic equation: s = s₀ + v*t + 0.5*a*t²
term1 = current_profit
term2 = velocity * dt
term3 = 0.5 * acceleration * dt**2
predicted_profit_raw = term1 + term2 + term3
# v0.2.1: ASYMMETRIC dampening - only dampen positive growth (optimism)
# Keep negative growth RAW (crash warnings must stay urgent!)
growth = term2 + term3
if growth > 0:
# Positive growth = over-optimism -> dampen it
growth_damped = growth * dampening
dampen_applied = True
else:
# Negative growth = crash warning -> keep RAW (urgent!)
growth_damped = growth
dampen_applied = False
predicted_profit = term1 + growth_damped
# DEBUG v0.2.1: Log calculation with asymmetric dampening (only for 60s)
if dt == 60:
dampen_str = f"× {dampening:.2f}" if dampen_applied else "× 1.00 (crash!)"
logger.info(
f"[TRAJ-CALC] {term1:.2f} + ({term2:.2f} + {term3:.2f}) {dampen_str} = "
f"{predicted_profit:.2f} (raw: {predicted_profit_raw:.2f}, regime: {regime})"
)
predictions.append(predicted_profit)
return predictions
def calculate_prediction_confidence(
self,
velocity_history: List[float],
acceleration_history: List[float]
) -> float:
"""
Hitung confidence level prediksi (0-1).
High confidence jika:
- Velocity stable (low variance)
- Acceleration consistent
- Sufficient data points
Args:
velocity_history: List of recent velocity values
acceleration_history: List of recent acceleration values
Returns:
Confidence score 0.0-1.0
"""
if len(velocity_history) < 3 or len(acceleration_history) < 3:
return 0.3 # Low confidence if insufficient data
# 1. Velocity stability (lower std = higher confidence)
vel_std = np.std(velocity_history[-5:])
vel_score = max(0, 1.0 - vel_std * 10) # Normalize
# 2. Acceleration consistency
accel_std = np.std(acceleration_history[-5:])
accel_score = max(0, 1.0 - accel_std * 100)
# 3. Data sufficiency bonus
data_score = min(len(velocity_history) / 20, 1.0) # Max at 20 samples
# Weighted average
confidence = vel_score * 0.4 + accel_score * 0.4 + data_score * 0.2
return min(max(confidence, 0.0), 1.0)
def should_hold_position(
self,
current_profit: float,
velocity: float,
acceleration: float,
min_target: float,
velocity_history: List[float] = None,
acceleration_history: List[float] = None,
regime: str = "normal"
) -> Tuple[bool, str, Dict[str, float]]:
"""
Rekomendasi apakah HOLD position berdasarkan prediksi (dengan regime dampening).
Args:
current_profit: Current profit ($)
velocity: Current velocity ($/s)
acceleration: Current acceleration ($/s²)
min_target: Minimum profit target ($)
velocity_history: Recent velocity values (optional)
acceleration_history: Recent acceleration values (optional)
regime: Market regime for dampening (v0.2.0)
Returns:
(should_hold, reason, predictions_dict)
Example:
>>> should_hold, reason, preds = predictor.should_hold_position(
... current_profit=0.05,
... velocity=0.1335,
... acceleration=0.0017,
... min_target=3.0,
... regime="volatile"
... )
>>> print(f"Hold: {should_hold}, Reason: {reason}")
Hold: True, Reason: Predicted $3.34 in 1min (target: $3.00)
"""
# v0.2.0: Predict 1m, 3m, 5m ahead with regime dampening
pred_1m, pred_3m, pred_5m = self.predict_future_profit(
current_profit, velocity, acceleration, regime=regime
)
# Calculate confidence (if history provided)
confidence = 1.0
if velocity_history and acceleration_history:
confidence = self.calculate_prediction_confidence(
velocity_history, acceleration_history
)
predictions = {
'pred_1m': pred_1m,
'pred_3m': pred_3m,
'pred_5m': pred_5m,
'confidence': confidence
}
# Decision logic
should_hold = False
reason = ""
# Check if low confidence - don't rely on predictions
if confidence < self.confidence_threshold:
reason = f"Low prediction confidence ({confidence:.0%}), use standard logic"
return False, reason, predictions
# HOLD if 1-minute prediction exceeds target significantly
if pred_1m > min_target * 2 and acceleration > 0:
should_hold = True
reason = f"Predicted ${pred_1m:.2f} in 1min (target: ${min_target:.2f}, conf: {confidence:.0%})"
# HOLD if strong acceleration even if current profit low
elif acceleration > 0.001 and velocity > 0.05 and pred_1m > min_target:
should_hold = True
reason = f"Strong acceleration ({acceleration:.4f}), pred ${pred_1m:.2f} > target"
# HOLD if recovering strongly (negative to positive trajectory)
elif current_profit < 0 and pred_1m > abs(current_profit) * 0.5:
should_hold = True
reason = f"Strong recovery trajectory: ${current_profit:.2f} -> ${pred_1m:.2f}"
# EXIT if prediction shows decline
elif pred_1m < current_profit * 0.8 and velocity < 0:
should_hold = False
reason = f"Declining trajectory: ${current_profit:.2f} -> ${pred_1m:.2f}"
else:
reason = f"Neutral prediction (1m: ${pred_1m:.2f})"
return should_hold, reason, predictions
def get_optimal_exit_time(
self,
current_profit: float,
velocity: float,
acceleration: float,
tp_target: float
) -> Tuple[float, int]:
"""
Estimasi waktu optimal untuk exit berdasarkan trajectory.
Args:
current_profit: Current profit
velocity: Current velocity
acceleration: Current acceleration
tp_target: Take profit target
Returns:
(peak_profit, time_to_peak_seconds)
Example:
>>> peak, time_to_peak = predictor.get_optimal_exit_time(
... current_profit=5.0,
... velocity=0.08,
... acceleration=-0.002, # Decelerating
... tp_target=10.0
... )
>>> print(f"Peak at ${peak:.2f} in {time_to_peak}s")
"""
# For parabolic motion with deceleration:
# Profit reaches peak when velocity = 0
# velocity(t) = v₀ + a*t = 0 -> t = -v₀/a
if acceleration >= 0:
# Still accelerating - no peak in near future
# Estimate based on reaching TP
if velocity > 0:
time_to_tp = (tp_target - current_profit) / velocity
return tp_target, int(time_to_tp)
else:
return current_profit, 0
# Decelerating (acceleration < 0)
time_to_peak = -velocity / acceleration # When velocity reaches 0
# Clamp to reasonable range (0-600 seconds = 10 minutes)
time_to_peak = max(0, min(time_to_peak, 600))
# Calculate peak profit
peak_profit = current_profit + velocity * time_to_peak + 0.5 * acceleration * time_to_peak**2
return peak_profit, int(time_to_peak)
if __name__ == "__main__":
# Test cases
predictor = TrajectoryPredictor()
# Test 1: Strong upward momentum (Trade #161613468 case)
print("=== Test 1: Strong Upward Momentum ===")
should_hold, reason, preds = predictor.should_hold_position(
current_profit=0.05,
velocity=0.1335,
acceleration=0.0017,
min_target=3.0
)
print(f"Should Hold: {should_hold}")
print(f"Reason: {reason}")
print(f"Predictions: 1m=${preds['pred_1m']:.2f}, 3m=${preds['pred_3m']:.2f}, 5m=${preds['pred_5m']:.2f}\n")
# Test 2: Declining trajectory
print("=== Test 2: Declining Trajectory ===")
should_hold, reason, preds = predictor.should_hold_position(
current_profit=5.0,
velocity=-0.05,
acceleration=-0.001,
min_target=3.0
)
print(f"Should Hold: {should_hold}")
print(f"Reason: {reason}")
print(f"Predictions: 1m=${preds['pred_1m']:.2f}\n")
# Test 3: Optimal exit time
print("=== Test 3: Optimal Exit Time ===")
peak, time_to_peak = predictor.get_optimal_exit_time(
current_profit=5.0,
velocity=0.08,
acceleration=-0.002,
tp_target=10.0
)
print(f"Peak Profit: ${peak:.2f}")
print(f"Time to Peak: {time_to_peak}s ({time_to_peak//60}m {time_to_peak%60}s)")