Currently reading Gappy's "Advanced Portfolio Management" and wanted to get a conversation going around how others think about factor models.
My simplistic thought: People find correlations between an asset and other baskets of assets and categorize that basket as a 'factor.' Then all performance of the asset is then benchmarked to those factors. Is this really just an extension of the idea around benchmarking an assets performance to the SP500, for example? Instead of just the SP500, make it 1) SP500, 2) small minus big market cap returns, and 3) returns of high book to market ratio minus returns of low.
Now obviously people have continually added more factors to the mix to explain returns. If I am a PM and discover that the market tends to rally on Tuesday's in October from 9-10 am when the low is under 30 degrees in NYC and I tell an academic about it will I start to see papers talking about the 9-10 AM Cold NYC October Tuesday factor?
Based on the most helpful WSO content, your understanding of factor models is on the right track. Factor models essentially decompose asset returns into components driven by systematic factors (like market risk, size, value, momentum, etc.) and idiosyncratic risk. The Fama-French 3-factor model, for instance, extends the CAPM by adding size (SMB: small minus big) and value (HML: high minus low book-to-market) factors to the market risk factor.
Your analogy of benchmarking to the S&P 500 and then expanding to additional factors is accurate. The idea is to explain as much of the asset's return as possible through these systematic factors. Over time, academics and practitioners have indeed added more factors to capture persistent anomalies or patterns in returns, such as momentum, quality, and low volatility.
As for your humorous example of the "9-10 AM Cold NYC October Tuesday factor," it highlights a key critique of factor proliferation: the risk of overfitting or data mining. Many factors identified in academic literature may not hold up out of sample or lack economic rationale. This is why robust testing and theoretical grounding are critical when introducing new factors.
If you're diving deeper into factor models, consider exploring the trade-offs between simplicity (e.g., fewer factors) and explanatory power, as well as the practical challenges of implementing multi-factor strategies in real-world portfolios.
Sources: Reflections from year 4 as an equity analyst, A Contrarian View - Raging Bull, https://www.wallstreetoasis.com/forum/hedge-fund/demystify-systematic-macro?customgpt=1, Reflections from year 4 as an equity analyst, Career in Passive Investment?
Tempore sequi aut odio accusamus libero ea nisi. Eveniet consequatur est delectus. Aut quis maiores perspiciatis consequatur veniam.
Excepturi alias tempora eveniet sed beatae quia totam. Commodi reiciendis voluptas nemo officiis. Non voluptatum sint quae. Unde beatae asperiores tenetur eos eveniet.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...