Why Characteristic Portfolios Do Not Require Full Investment

Summary

The document explains why the characteristic portfolio for a chosen asset characteristic is not constrained to have weights summing to one. In the cited construction, the portfolio minimizes variance subject to achieving a unit exposure to the characteristic. The question is whether it also needs the usual full investment condition for portfolio weights.

The response points to the book’s description of characteristic portfolios as potentially leveraged and containing both long and short positions. Their purpose is to isolate or scale exposure to a characteristic, rather than represent an ordinary fully invested portfolio. For example, if the characteristic values are small, substantial leverage may be needed to produce a characteristic-weighted exposure of one. Thus, a budget constraint would impose an additional restriction that is not part of the stated construction. The explanation is conceptual and relies on the book’s description; it does not derive the optimization solution or discuss implementation constraints such as leverage limits and transaction costs.

Key ideas

• A characteristic portfolio is defined by its exposure to a chosen characteristic, not by full investment.
• The construction minimizes portfolio variance subject to a unit characteristic exposure.
• Characteristic portfolios can contain long and short positions and may use leverage.
• Adding weights that sum to one would impose a separate budget constraint absent from the stated problem.
• Real implementation may still require explicit leverage and trading cost constraints.

Tags

# Active portfolio management - characteristic portfolios derivation


# Active portfolio management - characteristic portfolios derivation












In the book Active Portfolio Management by Grinold and Kahn, on page 30, when it derives the characteristic portfolio $h_a$ for some characteristic vector $a$, the problem is set up as $$\min h^TVh$$ s.t. $h^ta=1$

Why do we not need to add the constraint that $1^Th=1$ ($h$ should be a weight vector of a portfolio) here?

## Answer by Pleb (score 4, accepted)

https://quant.stackexchange.com/a/64349

Below proposition 1 (In the 2nd edition at p. 28?), at the beginning of the chapter, he specifically writes:

> Characteristic portfolios are not necessarily fully invested. They can include long and short positions and have significant leverage. Take the characteristic portfolio for earnings-to-price ratios. Since typical earnings-to-price ratios range roughly from 0.15 to 0, the characteristic portfolio will require leverage to generate a portfolio earnings-to-price ratio of 1. $[\ldots]$

In essence, he argues there is no need for the full investment constraint, hence $1^Th =1$ is excluded.

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.