Why Factor Z-Scores May Use Market-Capitalization Weights
Summary
The document explains why a multi-factor equity model might use market-cap-weighted means when converting factor values into Z-scores. In a universe with many small companies and a few very large ones, an unweighted cross-sectional regression can let the numerous smaller stocks dominate estimated factor returns. Weighting observations by capitalization gives larger companies more influence, so weighted Z-scores may approximate that emphasis when used in a factor model.
The choice depends on the intended model and benchmark. The answers connect capitalization weighting to market-cap-weighted beta and to matching institutional benchmark or investment styles for risk purposes. They also mention alternatives such as size-sorted factor portfolios, size indicators in a cross-sectional model, hierarchical models, and fundamental or free-float weighting. The document does not set a universal rule for when cap weighting is acceptable; its rationale is about the influence the model should assign to different firms and the characteristics it aims to represent.
Key ideas
- An unweighted cross-sectional regression can be influenced heavily by the large number of small-cap stocks.
- Market-cap weighting gives larger companies more influence on factor estimates.
- Cap-weighted Z-scores may approximate weighted least squares in a factor model.
- Alternative approaches include size-based factors, hierarchical models, and fundamental or free-float weights.
- The appropriate weighting scheme depends on the model’s purpose and target benchmark or investing style.
Tags
Full text
# When translating factors into normalized Z scores, why use cap-weighted means? # When translating factors into normalized Z scores, why use cap-weighted means? I've been working through this blog on multi-factor models and I noticed that, when translating factors into Z scores, the author uses cap-weighted means instead of regular means. Why would he do that? Are there certain markets where this practice is unacceptable? ## Answer by John (score 2, accepted) https://quant.stackexchange.com/a/34031 The distribution of equity market capitalizations is such that there are few large outliers (the Apples of the world), but then a lot more smaller companies. The consequence of this is that if you do a regression, such as a cross-sectional regression of the returns in period 1 against the book to price in period 0, then the coefficients will be dominated by the large number of small stocks. In other words, you would estimate a value factor return from period 0 to period 1 that's largely driven by the smaller-cap stocks. By contrast, a weighted least squares approach could be employed to put a greater emphasis on the larger market capitalization stocks. To the extent that Z-scores in a factor model are approximating this approach, the analysis holds. There are many other situations where models attempt to account for this. For instance, the Fama-French factors are constructed by splitting the universe into high and low market capitalization stocks. On a cross-sectional basis, one could use dummy variables to account for the different performance between the small and large stocks. More generally, a hierarchical model could be used. ## Answer by pyCthon (score 1) https://quant.stackexchange.com/a/34026 The choice affects how you construct beta among other things, and in the traditional CAPM sense beta is market cap weighted. There's also market cap squared or cubed weighting schemes. This allows the factor model to more accurately track institutional benchmarks or style of investing for risk purposes. There are also fundamental or free float weighting schemes.
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.