File size: 18,089 Bytes
f282470 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 | """Multi-Factor Scoring Engine v1.0 — Institutional-Grade Asset Scoring
Fuses 7 independent signals into a unified conviction score.
Based on: Gu et al. (2020) + Zuckerman (2021) Multi-Factor Equity Models
"""
import numpy as np
import pandas as pd
from typing import Dict, Optional
# Factor weights — calibrated for institutional portfolio allocation
# Weights sum to 1.0. Override for different strategies (value, momentum, etc.)
DEFAULT_WEIGHTS = {
'trend': 0.20, # Price momentum, SMA alignment
'momentum': 0.15, # RSI, MACD, stochastics
'volatility': 0.15, # Vol regime, vol-of-vol, realized vs implied
'fundamentals': 0.15, # Valuation, growth, quality metrics
'news': 0.15, # Sentiment, event risk, news flow
'options': 0.10, # Put/call ratio, gamma, IV skew
'macro': 0.10, # Dollar, rates, VIX, sector beta
}
# Category thresholds
CONVICTION_BANDS = {
'avoid': (0, 35), # Short / zero position
'neutral': (35, 55), # Watchlist / benchmark weight
'small': (55, 70), # 1/3 of target size
'moderate': (70, 85), # 2/3 of target size
'aggressive': (85, 100), # Full target size
}
class MultiFactorEngine:
"""Unified scoring from 7 orthogonal factors into a single conviction score."""
def __init__(self, weights: Optional[Dict[str, float]] = None):
self.weights = weights or dict(DEFAULT_WEIGHTS)
assert abs(sum(self.weights.values()) - 1.0) < 1e-6, "Weights must sum to 1.0"
# ── Factor 1: Trend ───────────────────────────────────────
@staticmethod
def trend_factor(price: float, sma20: float, sma50: float, sma200: float = None,
adx: float = None, high_52w: float = None, low_52w: float = None) -> float:
"""Score 0-100 based on price position vs moving averages.
Bullish when price > SMA20 > SMA50 with ADX > 25 confirming.
"""
score = 50.0
if price > sma20: score += 15
if price > sma50: score += 15
if sma20 > sma50: score += 10
if sma200 and price > sma200: score += 10
if high_52w and low_52w and high_52w > low_52w:
score += 10 * ((price - low_52w) / (high_52w - low_52w + 1e-10))
if adx and adx > 25:
score += 10
elif adx and adx > 40:
score += 15
return max(0, min(100, score))
# ── Factor 2: Momentum ────────────────────────────────────
@staticmethod
def momentum_factor(rsi: float, macd_hist: float, obv_slope: float = None,
rsi_divergence: float = None) -> float:
"""Score 0-100. Momentum peaking at RSI 50-70 range.
Bearish divergence (price up, RSI down) reduces score.
"""
score = 50.0
if rsi < 30: score += 15 # Oversold bounce
elif 30 <= rsi <= 45: score += 5
elif 45 < rsi <= 65: score += 10 # Healthy momentum
elif 65 < rsi <= 75: score += 3 # Still strong but caution
elif rsi > 75: score -= 15 # Overbought
if macd_hist > 0: score += 10
if macd_hist > 0.5: score += 10
if macd_hist < 0: score -= 10
if obv_slope and obv_slope > 0: score += 10
if obv_slope and obv_slope < 0: score -= 10
if rsi_divergence and rsi_divergence < 0: score -= 15 # Bearish div
if rsi_divergence and rsi_divergence > 0: score += 10 # Bullish div
return max(0, min(100, score))
# ── Factor 3: Volatility ─────────────────────────────────
@staticmethod
def volatility_factor(hv: float, iv: float = None, vol_regime: str = 'normal',
skew: float = None, vix: float = None) -> float:
"""Score 0-100. Low vol regime = bullish. High vol with declining trend = bullish.
IV < HV (cheap options) = bullish. Rising vol = bearish.
"""
score = 50.0
if hv < 0.15: score += 15
elif hv < 0.25: score += 5
elif hv > 0.40: score -= 15
elif hv > 0.35: score -= 5
if iv and hv and iv < hv:
score += 10 # Cheap implied vol
if iv and hv and iv > hv * 1.5:
score -= 15 # Expensive options, fear premium
if vol_regime == 'low': score += 15
if vol_regime == 'declining': score += 10
if vol_regime == 'spiking': score -= 20
if vol_regime == 'high': score -= 10
if skew and skew < 0: score += 5 # Put skew mild
if skew and skew < -0.5: score -= 10 # Extreme fear
if vix and vix < 15: score += 10
if vix and vix > 30: score -= 15
return max(0, min(100, score))
# ── Factor 4: Fundamentals ────────────────────────────────
@staticmethod
def fundamentals_factor(pe: float = None, peg: float = None, ps: float = None,
pb: float = None, roe: float = None, debt_equity: float = None,
fcf_yield: float = None, growth_5y: float = None,
sector_pe: float = None) -> float:
"""Score 0-100 based on valuation + quality. Lower PEG, higher ROE = better.
FCF yield > 5% is strong signal. Relative to sector PE.
"""
score = 50.0
# Valuation
if pe:
if pe < 12: score += 15
elif pe < 18: score += 10
elif pe < 25: score += 3
elif pe > 35: score -= 10
elif pe > 50: score -= 15
if sector_pe and pe < sector_pe * 0.8: score += 10
if peg:
if peg < 0.8: score += 15
elif peg < 1.0: score += 10
elif peg < 1.5: score += 3
elif peg > 2.0: score -= 15
if ps:
if ps < 1: score += 10
elif ps > 10: score -= 10
if pb:
if pb < 1.5: score += 10
elif pb > 5: score -= 10
# Quality
if roe and roe > 0.15: score += 10
if roe and roe > 0.25: score += 5
if roe and roe < 0.05: score -= 10
if debt_equity:
if debt_equity < 0.5: score += 10
elif debt_equity > 2.0: score -= 10
if fcf_yield:
if fcf_yield > 0.06: score += 15
elif fcf_yield > 0.03: score += 10
elif fcf_yield < 0: score -= 15
if growth_5y:
if growth_5y > 0.20: score += 10
elif growth_5y > 0.15: score += 5
elif growth_5y < 0: score -= 15
return max(0, min(100, score))
# ── Factor 5: News ────────────────────────────────────────
@staticmethod
def news_factor(sentiment_score: float, news_volume: float = None,
event_risk: str = 'none', event_date: str = None) -> float:
"""Score 0-100. Sentiment 0.0-1.0 (bearish-bullish). Event risk overrides.
sentiment_score: 0.0=very bearish, 0.5=neutral, 1.0=very bullish
news_volume: articles per day, higher = more signal confidence
event_risk: 'none', 'earnings_today', 'earnings_week', 'fed_today',
'macro_today', 'lawsuit', 'merger', 'dividend'
"""
score = 50.0
# Base sentiment signal
if sentiment_score < 0.2: score -= 25
elif sentiment_score < 0.35: score -= 15
elif sentiment_score < 0.45: score -= 5
elif 0.45 <= sentiment_score <= 0.55: score += 0
elif sentiment_score > 0.7: score += 25
elif sentiment_score > 0.55: score += 15
elif sentiment_score > 0.5: score += 5
# Volume confidence boost
if news_volume:
if news_volume > 50: score += 5
if news_volume > 200: score += 5 # High coverage = signal confidence
# Event risk override
event_override = {
'none': 0,
'earnings_week': -5, # Reduce size before earnings
'earnings_today': -20, # Major uncertainty
'fed_today': -15,
'macro_today': -10,
'lawsuit': -25,
'merger': 10, # Merger arb or acquisition premium
'dividend': 3,
'split': 5,
'buyback': 8,
}
if event_risk in event_override:
score += event_override[event_risk]
return max(0, min(100, score))
# ── Factor 6: Options Flow ───────────────────────────────
@staticmethod
def options_factor(pcr: float = None, unusual_volume: float = None,
gamma_exposure: float = None, iv_skew: float = None,
open_interest_change: float = None, max_pain: float = None,
current_price: float = None) -> float:
"""Score 0-100 based on options market microstructure signals.
pcr: put/call ratio < 0.5 bullish, > 1.2 bearish
unusual_volume: ratio vs 20d avg > 2 = significant
gamma_exposure: positive gamma = sticky price, negative = magnetic pin
iv_skew: steep put skew = fear, flat = complacent
open_interest_change: rising OI with rising price = conviction
max_pain: price tends toward max pain on expiry
"""
score = 50.0
if pcr:
if pcr < 0.5: score += 20 # Extreme call buying
elif pcr < 0.7: score += 10
elif pcr > 1.2: score -= 20 # Extreme put buying
elif pcr > 1.0: score -= 10
elif pcr > 0.85: score -= 3 # Mild put bias
if unusual_volume:
if unusual_volume > 5: score += 15 # Massive flow
elif unusual_volume > 2: score += 8 # Notable flow
elif unusual_volume < 0.5: score -= 5 # Dead options
if gamma_exposure:
if gamma_exposure > 0: score += 5 # Gamma long = volatility support
elif gamma_exposure < -5: score -= 15 # Gamma short = volatility risk
if iv_skew:
if iv_skew < -0.3: score -= 10 # Fear
if iv_skew < -0.5: score -= 15 # Extreme fear
if iv_skew > 0.1: score += 5 # Call skew (rare, bullish)
if open_interest_change and open_interest_change > 0.3:
score += 10 # Fresh conviction building
if max_pain and current_price:
if current_price < max_pain * 0.97: score += 5 # Room to max pain (up)
if current_price > max_pain * 1.03: score -= 5 # Above max pain (mean revert)
return max(0, min(100, score))
# ── Factor 7: Macro ───────────────────────────────────────
@staticmethod
def macro_factor(vix: float = None, dxy_change: float = None,
yield_10y: float = None, yield_2y: float = None,
sector_beta: float = None, cpi_surprise: float = None,
fed_meeting_days: int = None) -> float:
"""Score 0-100. Macro tailwinds or headwinds for risk assets.
vix: volatility index
dxy_change: 20d % change in dollar index
yield_10y / yield_2y: treasury yields
sector_beta: stock's beta to rates (tech > 1, utilities < 0.5)
cpi_surprise: actual - expected, > 0 = hawkish
fed_meeting_days: days until next FOMC
"""
score = 50.0
# VIX regime
if vix:
if vix < 15: score += 15
elif vix < 20: score += 5
elif vix > 30: score -= 20
elif vix > 25: score -= 10
elif vix > 20: score -= 5
# Dollar strength — bad for exporters, good for domestic
if dxy_change:
if dxy_change > 0.05: score -= 10 # Strong dollar
if dxy_change < -0.03: score += 5 # Weak dollar
# Yield curve
if yield_10y and yield_2y:
spread = yield_10y - yield_2y
if spread < -0.5: score -= 15 # Deep inversion = recession
elif spread < 0: score -= 10 # Inverted
elif spread > 1.0: score += 10 # Steep = healthy
# Rising rates hurt rate-sensitive stocks
if yield_10y and sector_beta:
if yield_10y > 0.05 and sector_beta > 1.0:
score -= 15 # High rates + high beta = double hurt
elif yield_10y < 0.03 and sector_beta > 1.0:
score += 10 # Low rates + high beta = double boost
# CPI surprise
if cpi_surprise:
if cpi_surprise > 0.5: score -= 15 # Hot inflation = hawkish Fed
if cpi_surprise < -0.3: score += 10 # Cold inflation = dovish
# Fed meeting proximity
if fed_meeting_days is not None:
if fed_meeting_days <= 3: score -= 10 # Imminent uncertainty
if fed_meeting_days == 0: score -= 20 # Meeting day
return max(0, min(100, score))
# ── Master Scoring ────────────────────────────────────────
def score(self, factors: Dict[str, float], verbose: bool = False) -> Dict:
"""Compute unified conviction score from individual factor scores.
Args:
factors: dict with keys matching self.weights
e.g. {'trend': 75, 'momentum': 45, 'volatility': 60,
'fundamentals': 80, 'news': 65, 'options': 70, 'macro': 55}
verbose: return full breakdown
Returns:
Dict with score, band, and optionally per-factor contributions
"""
total = 0.0
contributions = {}
for key, weight in self.weights.items():
val = factors.get(key, 50.0) # Default neutral if missing
contrib = val * weight
contributions[key] = {
'score': val,
'weight': weight,
'contribution': contrib,
}
total += contrib
total = max(0, min(100, total))
# Determine band
band = 'neutral'
for name, (lo, hi) in CONVICTION_BANDS.items():
if lo <= total < hi:
band = name
break
if total >= 85:
band = 'aggressive'
# Position sizing based on band
sizing = {
'avoid': 0.00,
'neutral': 0.00,
'small': 0.33,
'moderate': 0.67,
'aggressive': 1.00,
}
result = {
'conviction_score': round(total, 2),
'band': band,
'target_exposure': sizing.get(band, 0),
'direction': 'long' if total > 55 else 'short' if total < 35 else 'neutral',
}
if verbose:
result['contributions'] = contributions
result['weights'] = dict(self.weights)
return result
def score_from_dataframe(self, df: pd.DataFrame,
fundamentals: Optional[Dict] = None,
news: Optional[Dict] = None,
options: Optional[Dict] = None,
macro: Optional[Dict] = None,
verbose: bool = False) -> Dict:
"""Compute multi-factor score from price DataFrame + optional overlays."""
l = df.iloc[-1]
p = df.iloc[-2] if len(df) > 1 else l
# Trend
trend = self.trend_factor(
price=l['Close'], sma20=l.get('SMA20', l['Close']),
sma50=l.get('SMA50', l['Close']),
high_52w=df['High'].max(), low_52w=df['Low'].min()
)
# Momentum
macd_hist = l.get('MACD', 0) - l.get('MACDS', 0)
momentum = self.momentum_factor(rsi=l.get('RSI', 50), macd_hist=macd_hist)
# Volatility
hv = df['Ret'].dropna().std() * np.sqrt(252)
volatility = self.volatility_factor(hv=hv)
# Fundamentals
fund = 50.0
if fundamentals:
fund = self.fundamentals_factor(**fundamentals)
# News
news_score = 50.0
if news:
news_score = self.news_factor(**news)
# Options
opt = 50.0
if options:
opt = self.options_factor(**options)
# Macro
macro_score = 50.0
if macro:
macro_score = self.macro_factor(**macro)
factors = {
'trend': trend,
'momentum': momentum,
'volatility': volatility,
'fundamentals': fund,
'news': news_score,
'options': opt,
'macro': macro_score,
}
return self.score(factors, verbose=verbose)
if __name__ == '__main__':
# Example: Microsoft Corporation with user's actual metrics
engine = MultiFactorEngine()
factors = {
'trend': 85.0, # Price > SMA20 > SMA50
'momentum': 45.0, # RSI 48.9, MACD bearish crossover
'volatility': 30.0, # High vol (29.4%), negative Sharpe
'fundamentals': 80.0, # MSFT: great fundamentals
'news': 50.0, # Neutral sentiment
'options': 55.0, # Mild activity
'macro': 40.0, # Rising yields, neutral VIX
}
result = engine.score(factors, verbose=True)
print(f"Conviction Score: {result['conviction_score']}/100")
print(f"Band: {result['band'].upper()}")
print(f"Direction: {result['direction'].upper()}")
print(f"Target Exposure: {result['target_exposure']*100:.0f}%")
print("\nPer-factor breakdown:")
for k, v in result['contributions'].items():
print(f" {k:12s}: {v['score']:5.1f} × {v['weight']:.2f} = {v['contribution']:.1f}")
|