Personal Investments • Total International Stock Index Fund vs. U.S. Small Cap Value for Diversification

Just my two cents, but I mostly think of myself as just a value investor, and I think of most other factors as relevant to the extent I think they may help an identify a portfolio that really has the sorts of value characteristics I want in substance. Like I don't necessarily think of Small as a really independent thing, I more think that Small and Value as defined in certain ways together might better capture what I might call a more meta sense of value. Other factors, to the extent I really think about them at all, I similarly see as really most meaningful in the sense of contributing to this meta sense of value.

And that's good, because there is a lot of that going on in practice. Like these days, CRSP's way of identifying Value for its various Value indices actually uses five different factors in what it calls a multi-factor model:

https://www.crsp.org/wp-content/uploads ... _Guide.pdf

1. Book-to-Price Ratio (BP)
2. Future Earnings-to-Price Ratio (FEP)
3. Historical Earnings-to-Price Ratio (HEP)
4. Dividend-to-Price Ratio (DP)
5. Sales-to-Price Ratio (SP)

This could be seen as incorporating elements of Profitability and Quality and so on, not just "pure" Value. And it is fine with me CRSP puts that all under the umbrella of Value, because again that is pretty much how I think anyway.

If you are curious, this is what CRSP does with all that:

1. For each company, determine values for the various style factors.
2. Winsorize the data for each style factor at the 5th and 95th percentiles.
3. Calculate each style factor’s cross-sectional mean and cross-sectional standard deviation.
4. Scale each style factor for each company by finding the factor’s z-score using the following formula:
z score = company factor - cross sectional mean for factor/cross sectional standard deviation for the factor

The z-score factors are denoted by the presence of a “z” before the factor acronym (e.g. zBP).
. . .
CRSP’s model first combines the future and historical earnings-to-price ratios (FEP and HEP) into a single earnings-to-price factor (EP), which is then combined with the book-to-price ratio (BP) to create a primary value super factor (V1). The sales-to-price ratio (SP) and dividend-to-price ratio (DP) are then combined to create a secondary value super factor (V2). Merging the two value super factors together creates a single composite value score (V), as defined below:

V = (2 ⁄ 3 × V1) + (1 ⁄ 3 × V2), where:
V1 = (2 ⁄ 3 × zEP) + (1 ⁄ 3 × zBP), where zEP = (2 ⁄ 3 × zFEP) + (1 ⁄ 3 × zHEP)
V2 = (2 ⁄ 3 × zSP) + (1 ⁄ 3 × zDP)

So that's "fun". Is that the best definition of value? Who knows. But I get the general idea, I am cool with it, so don't have a problem with value ETFs using CRSP's indices.

Statistics: Posted by NiceUnparticularMan — Sun Oct 19, 2025 7:54 pm — Replies 81 — Views 6060

---