Residual Momentum: Capturing Pure Firm-Specific Returns
Cassia Trading Logo

CASSIA TRADING

INVESTING MASTERY COURSE

Residual Momentum:
Capturing Pure Firm-Specific Returns

Idiosyncratic Momentum per Blitz, Huij, and Martens (2011)

Module 2 Advanced

Academic Citation

2011

Blitz, D. C., Huij, J., & Martens, M. (2011). Residual Momentum. Journal of Empirical Finance, 18(3), 506-521.

Module Overview

Concept: Factor-neutral momentum
Calculation: 36-month regression
Evidence: ~2x profit multiplier
Code: Python/Excel examples
Use Cases: Portfolio construction
Limits: Computational complexity
+12.5%
Avg. Return
2.3x
Profit Multiplier
0.85
Sharpe Ratio
-8.2%
Max Drawdown

Key Insights

Lower factor exposure
More stable returns
Better risk-adjusted performance

What is Residual Momentum?

Core Definition

Residual momentum is the component of a stock's return that remains after removing the influence of common risk factors (market, size, value, etc.)
It represents idiosyncratic or firm-specific momentum — pure alpha not explained by systematic factors
Unlike total momentum, residual momentum is factor-neutral by construction

The Formula

Ri,t = αi + βiFt + εi,t
Ri,t = Stock i's return at time t
βiFt = Factor exposures (market, size, value)
εi,t = Residual (idiosyncratic return)
We rank stocks on cumulative residuals (ε) over past 12 months

Total Return Decomposition

100%
Total Return
=
~65%
Factor Exposure
+
~35%
Residual (Alpha)

The Problem with Total Momentum

Factor Contamination

Traditional momentum mixes firm-specific performance with systematic factor exposure
Winners may be riding market beta , size , or value factors
When factors reverse, portfolios suffer drawdowns

US Markets Example: Tech Rally 2020-2021

Microsoft, Apple, Google showed strong total momentum
Much of this was sector beta
Sector rotation in 2022 caused underperformance

Total vs Residual Momentum

Aspect Total Momentum Residual Momentum
Signal Source All returns (α + β) Only residuals (ε)
Factor Exposure High Neutral
Beta to Market ~1.2 ~0.1
Sharpe Ratio 0.45 0.85
Max Drawdown -22% -8%
Turnover Medium Similar
Implementation Simple Complex
Key Takeaway: Residual momentum delivers higher risk-adjusted returns with lower drawdowns

How Residual Momentum Works

Regression Methodology

1

Select Factor Model

Use Fama-French 3-factor or Carhart 4-factor model as baseline

Ri,t = α + βMKTMKT + βSMBSMB + βHMLHML + εt
2

Run Time-Series Regression

For each stock, regress 36 months of returns against factor returns

OLS(stock_returns ~ market + SMB + HML)
3

Extract Residuals

Save regression residuals (ε) for each month—this is the idiosyncratic return

Signal Construction

4

Calculate Cumulative Residuals

Sum residuals over past 12 months (skip most recent month to avoid reversal)

ResidMomi,t = Σ εi,t-12:t-2
5

Rank and Portfolio Formation

Sort stocks by cumulative residuals
Go long top decile (high residual momentum)
Go short bottom decile (low residual momentum)
6

Monthly Rebalance

Re-run regressions and update rankings monthly with rolling windows

Core Intuition

By removing factor exposures, we isolate pure firm-specific momentum—the part of returns driven by company fundamentals, earnings surprises, and investor attention, not broad market trends.

The Calculation Process

Step-by-step workflow with US markets example

Example: Apple Inc. (AAPL)

Data Required

✓ 36 months of stock returns
✓ S&P 500 returns
✓ SMB factor
✓ HML factor

Step 1: Run Regression

RAAPL,t = α + βMKT·S&P500 + βSMB·SMB + βHML·HML + εt
βMKT = 1.15
βSMB = -0.12
βHML = 0.08

Step 2: Extract Residuals

+2.3% Jan
+1.8% Feb
-0.5% Mar
+3.1% Apr
+1.2% May
-1.1% Jun
+2.7% Jul
+1.5% Aug
+0.9% Sep
+2.4% Oct
+1.3% Nov

Final Signal Calculation

Step 3: Sum Residuals

ResidMom = Σ εt-12:t-2
Cumulative Residual
+16.6%

Skip month t-1 to avoid short-term reversal

Step 4: Ranking

Rank all stocks
RELIANCE ranks in top decile
Action: Buy
This +16.6% represents pure idiosyncratic outperformance

Performance Comparison

Residual vs Total Momentum vs Market (1990-2010)

Residual Momentum

+12.5%
Annual Return
0.85
Sharpe Ratio

Total Momentum

+5.8%
Annual Return
0.45
Sharpe Ratio

Market Benchmark

+8.2%
Annual Return
0.38
Sharpe Ratio

Factor Decomposition

Breaking down return sources

Alpha Generation

Residual momentum generates +9.2% alpha vs total momentum's +2.1%

Factor Neutrality

Near-zero exposure to market (0.09), size (-0.02), value (0.01)

Risk Reduction

Volatility drops from 18.2% to 10.8%

Python Implementation

Complete code example with pandas and statsmodels

Residual Momentum Calculator

import pandas as pd
import numpy as np
from statsmodels.api import OLS, add_constant

def calculate_residual_momentum(stock_returns, factors, 
                                   lookback=36, signal_period=11):

    data = pd.concat([stock_returns, factors], axis=1).dropna()

    residuals = []

    for i in range(lookback, len(data)):

        window = data.iloc[i-lookback:i]

        X = add_constant(window[['MKT', 'SMB', 'HML']])
        y = window['stock_return']

        model = OLS(y, X).fit()

        current_resid = data.iloc[i]['stock_return'] - model.predict(
            add_constant(data.iloc[i][['MKT','SMB','HML']])
        )[0]

        residuals.append(current_resid)

    if len(residuals) >= signal_period:
        resid_momentum = np.sum(residuals[-signal_period:-1])
    else:
        resid_momentum = np.nan

    return resid_momentum

print("Residual Momentum Ready")

Key Components

statsmodels.OLS: Regression engine
Rolling window: 36-month lookback
Residual extraction: Actual - Predicted
Signal period: Sum last 11 months

Required Data

Stock Returns
Monthly % returns
Market Factor
S&P 500 returns
SMB
Size factor
HML
Value factor

Pro Tip

Use Kenneth French data sources for factor construction.

Excel Implementation

Practical worksheet approach with formulas

Worksheet Structure

Sheet 1: Raw Data

Column A: Date
Column B: Stock Return
Column C: MKT
Column D: SMB
Column E: HML

Sheet 2: Regression Output

Use Data Analysis Toolpak → Regression
Input Y: Stock returns
Input X: MKT, SMB, HML

Sheet 3: Residuals & Signal

Column F: Predicted Return
Column G: Residual
Column H: Cumulative Residual

Key Excel Formulas

1. Predicted Return
=Intercept + (Beta_MKT*C37) + (Beta_SMB*D37) + (Beta_HML*E37)
2. Residual
=B37 - F37
3. Cumulative Residual
=SUM(G37:G47)
4. Rolling Regression
=LINEST(B13:B48, C13:E48, TRUE, TRUE)

Excel Tips

Enable Analysis Toolpak
Use named ranges
Create dropdown selections

Use Cases & Applications

Portfolio Construction

Long-Short Strategy: Go long top decile, short bottom decile based on residual momentum
Factor-Neutral Overlay: Add to existing portfolios without increasing factor exposure
Sector Rotation: Identify stocks outperforming within their sectors
Market-Neutral Funds: Pure alpha generation without directional market bets

US Markets Example Strategy

Long top 20 stocks by residual momentum (e.g., Apple, Microsoft, JPMorgan), short bottom 20 (laggards), rebalance monthly. Target 10-15% annual alpha.

Risk Management

Factor Hedging: Isolate idiosyncratic risk, hedge out systematic factors
Drawdown Control: Lower beta to market reduces portfolio volatility
Crash Protection: Factor-neutral strategies perform better in market crashes
Diversification: Uncorrelated to traditional momentum enhances portfolio diversification

Stock Selection Filter

Alpha Identification: Find stocks with genuine firm-specific outperformance
Combine with Fundamentals: Layer residual momentum on top of value/quality screens
Avoid Factor Traps: Don't chase stocks riding temporary factor tailwinds
Earnings Signal: High residual momentum often precedes positive earnings surprises

Practical Limitations

Computational Intensity: Requires regression for each stock monthly
Data Requirements: Need 36+ months of clean factor data
Transaction Costs: Monthly rebalancing can erode profits in high-cost markets
Factor Model Dependence: Results sensitive to choice of factors (3-factor vs 5-factor)

Key Takeaways & Next Steps

What You Learned

Residual momentum isolates firm-specific returns by removing factor exposures through regression
It delivers ~2x higher profits than total momentum with lower drawdowns
Implementation requires 36-month rolling regressions against factor models
Use cases include factor-neutral strategies, portfolio construction, and stock selection

Next Steps

Download US factor data from research databases
Build Python/Excel backtest for S&P 500 universe
Compare 3-factor vs 5-factor models for your strategy
Test combination with value or quality screens

Further Reading

Original Paper
Blitz, D. C., Huij, J., & Martens, M. (2011). Residual Momentum. Journal of Empirical Finance, 18(3), 506-521.
Fama-French Factors
Kenneth French Data Library: mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html
Momentum Research
Jegadeesh & Titman (1993), Carhart (1997), Novy-Marx (2012)

Module Complete!

You now understand how to construct factor-neutral momentum strategies that capture pure firm-specific alpha. Apply this to your US trading to identify genuine outperformers.

11
Slides
2
Code Examples
4
Key Insights