Executive Summary

This report presents a systematic multi-factor equity strategy constructed and backtested on a 50-stock universe drawn from the S&P 500. The strategy selects stocks monthly using three well-documented return premia — Momentum, Value, and Quality — combined into a composite z-score ranking. The top 15 stocks are held in an equal-weighted portfolio, rebalanced monthly.

Metric Strategy S&P 500
Annualised Return 18.4% 15.2%
Sharpe Ratio 0.58 0.75
Annualised Alpha 6.21% —
Beta 0.96 1.00
Max Drawdown 95% 98.3%
Monthly Win Rate 50% —

Key finding: The strategy generated 6.21% annualised alpha with a near-market beta of 0.96. Excess returns were driven by factor selection skill rather than elevated market risk. The lower Sharpe ratio reflects higher absolute volatility — a known characteristic of momentum-tilted strategies, manageable through position sizing in a live portfolio.

Investment Rationale

Why Factor Investing?

Factor investing is a systematic approach to portfolio construction that selects securities based on specific, measurable characteristics persistently associated with excess returns. Unlike discretionary stock picking, factor strategies are rules-based, scalable, and free from emotional bias.

The three factors in this strategy are among the most extensively documented in finance literature:

  • Momentum — Jegadeesh & Titman (1993) showed stocks outperforming over the past 12 months tend to continue outperforming. Markets are slow to fully price good news, creating a persistent short-to-medium term trend.

  • Value — De Bondt & Thaler (1985) identified a long-term reversal effect: stocks significantly underperforming over 3–5 years tend to subsequently recover. Markets systematically overpunish losers relative to intrinsic worth.

  • Quality — Companies with stable, consistent risk-adjusted earnings are more resilient across economic cycles. A rolling Sharpe ratio proxies for business model quality — rewarding consistency over raw magnitude.

Why Combine All Three?

Each factor experiences extended underperformance in isolation. Momentum crashes in sharp market reversals. Value significantly underperformed during the 2010–2020 growth cycle. Combining factors with low correlation reduces the probability of simultaneous failure — factor diversification directly analogous to asset class diversification in traditional portfolio management.

Methodology

Universe

The strategy operates on a 50-stock subset of the S&P 500, drawn equally across Technology, Financials, Healthcare, Energy, and Consumer & Industrials. All data is sourced from Yahoo Finance via the tidyquant R package using adjusted closing prices to account for dividends and stock splits.

Factor Construction

Factor Measure Construction Logic
Momentum 12-month return, lagged 1 month Lag avoids short-term reversal effect
Value Negative 36-month return Contrarian — lower long-term = higher score
Quality Rolling 24-month Sharpe ratio mean(return) / sd(return) over 24 months

Score Combination

In each month t, raw factor scores are cross-sectionally z-scored across all 50 stocks before combination:

\[z_i = \frac{x_i - \mu}{\sigma} \qquad \text{Composite}_i = \frac{z_{\text{mom}} + z_{\text{val}} + z_{\text{qual}}}{3}\]

The top 15 stocks by composite score are selected and held equal-weighted, rebalanced on the first trading day of each month.

Backtest Results

Cumulative Performance

£1 invested at inception grew to £6.2325103^{9} by December 2024, versus £1.7259731^{8} for the benchmark — 3.1 percentage points of annual outperformance driven by factor skill rather than market risk (beta: 0.96).

Drawdown Analysis

The strategy’s maximum drawdown of 95% reflects concentrated equity exposure during broad market dislocations — notably the COVID-19 selloff of March 2020.

Annual Return Breakdown

Year Strategy Benchmark Excess Return
2016 766.3% 1 077.1% -310.9%
2017 6 783.9% 1 803.9% 4 980.0%
2018 -42.7% -50.4% 7.7%
2019 2 517.1% 5 791.0% -3 273.9%
2020 7 079.0% 1 148.9% 5 930.1%
2021 1 476.9% 4 317.1% -2 840.2%
2022 95.9% -95.1% 191.0%
2023 316.9% 3 170.7% -2 853.8%
2024 7 440.5% 2 861.2% 4 579.3%

Current Portfolio Holdings

Rank Ticker Momentum Score Value Score Quality Score Composite Score
1 META 1.09 -0.23 2.30 1.05
2 WMT 1.54 -0.66 1.76 0.88
3 AMZN 0.31 0.46 1.54 0.77
4 NVDA 4.61 -4.72 2.41 0.76
5 TSLA 0.37 0.86 0.64 0.62
6 ISRG 1.03 0.07 0.68 0.59
7 GOOGL -0.16 0.58 1.19 0.54
8 AXP 1.15 -0.57 1.01 0.53
9 JPM 0.71 -0.24 0.99 0.49
10 GS 1.03 -0.24 0.61 0.47
11 AAPL -0.08 0.30 1.17 0.46
12 ORCL 1.53 -0.99 0.71 0.42
13 CRM -0.02 0.38 0.85 0.40
14 WFC 0.97 -0.25 0.46 0.39
15 BAC 0.52 0.61 0.00 0.38

Risk & Limitations

A rigorous evaluation of any backtested strategy requires honest acknowledgement of its limitations. The following risks are material and should be understood before drawing conclusions from these results.

Overfitting. Factor definitions and parameter choices were selected with knowledge of the historical period. While the three factors are grounded in decades of academic evidence, the specific window lengths have not been validated on a held-out sample.

Survivorship bias. The 50-stock universe consists of current S&P 500 constituents. Stocks delisted between 2015 and 2024 are absent, likely overstating returns modestly.

Transaction costs. Monthly rebalancing incurs trading costs not captured here. Realistic round-trip costs of 5–10 basis points per trade reduce annualised returns by approximately 0.5–1.0%.

Market regime dependency. Outperformance was concentrated in specific periods. Performance in alternative macro regimes may differ materially.

What I Would Do Differently With More Resources

  1. Expand to the full S&P 500 for more robust cross-sectional signals
  2. Incorporate fundamental data (P/E, ROE, debt ratios) for cleaner Value and Quality factors
  3. Add sector neutrality constraints to prevent unintended concentration
  4. Test out-of-sample on MSCI Europe and MSCI Asia for cross-market robustness
  5. Implement a volatility-scaling overlay to reduce exposure in high-volatility regimes

Conclusion

This project demonstrates the construction and evaluation of a systematic multi-factor equity strategy in R. The combination of Momentum, Value, and Quality factors produced a 18.4% annualised return over 2015–2024 — outperforming the S&P 500 by 3.1 percentage points annually — with a regression-derived alpha of 6.21%.

The results reinforce a core principle of systematic asset management: that disciplined, rules-based processes can generate persistent excess returns by exploiting well-documented behavioural and structural market anomalies — without requiring individual stock selection skill.

Built in R · tidyquant · PerformanceAnalytics · ggplot2 · kableExtra · Data: Yahoo Finance · Backtest: January 2015 – December 2024 · University of Warwick, MORSE · April 2026