The first quarter of 2025 demonstrated the continued evolution of factor-based investing, with traditional factor premiums showing mixed performance while alternative data-driven factors gained prominence. Our multi-factor model review analyzes the key drivers of alpha generation and risk decomposition.
Factor Performance Overview
The Value factor delivered its strongest quarterly performance in three years, generating 3.8% excess return as undervalued sectors benefited from policy support and mean reversion. The Momentum factor continued its winning streak with 2.9% alpha, particularly strong in the technology and consumer discretionary sectors. Quality factor returned 2.1%, with high-ROE and low-leverage companies outperforming in a rising rate environment.
The Volatility factor provided 1.5% excess return, as low-volatility strategies benefited from market uncertainty. The Size factor underperformed with -0.8% alpha, as large-cap stocks dominated the rally. Our composite multi-factor model, equally weighting these five factors, generated a total excess return of 1.9% with a Sharpe ratio of 1.35.
Alpha Attribution Analysis
Stock selection contributed 65% of total alpha, while sector allocation contributed 35%. Within stock selection, the consumer discretionary and financial sectors were the largest alpha contributors. The energy sector detracted from performance due to volatile oil prices disrupting factor signals.
Risk Decomposition
Total portfolio risk stood at 11.2%, with systematic risk contributing 8.1% and idiosyncratic risk 3.1%. The top risk contributors were market beta (45%), sector exposure (22%), and factor exposure (18%). Currency risk for international portfolios added 1.5% to total risk, reflecting dollar strength.
Emerging Factor Trends
ESG factors showed increasing significance, with the environmental score contributing 0.8% positive alpha. Sentiment analysis factors derived from news and social media data added 1.1% alpha, suggesting growing importance of alternative data sources. Machine learning-based factor timing models improved factor allocation efficiency by 0.6%.
