Portfolio concentration: holdings, largest weight and HHI

Concentration describes how much an allocation is clustered in a small number of holdings. Two portfolios with the same security count can distribute their value very differently.

The count, largest weight and a combined weight statistic convey different information. A low concentration number alone does not establish low volatility or shallow losses.

Consider a fictional teaching example with two securities weighted 50/50, and another weighted 90/10. Both have two holdings, but their largest weights are 50% and 90%. Neither allocation is an investment recommendation.

Using fractional weights, HHI is the sum of squared weights. The first gives 0.5²+0.5² = 0.50; the second gives 0.9²+0.1² = 0.82. The latter has more uneven weights despite having the same count.

For nonnegative weights normalized to sum to one, equal weights across n holdings give HHI = 1/n; one holding gives 1. These are arithmetic properties, not safety thresholds.

Squaring percentage numbers directly gives 50²+50² = 5,000 for the same equal allocation. That scale differs from 0.50. Check units and the representation of weights before comparing reports.

In a fictional account worth 1,000,000 yen, suppose cash is 500,000 and A and B each represent 250,000. Each security is 25% of total account value, or 50% when normalized within equities.

The security-only squared sum on a total-account basis is 0.125, equity-normalized HHI is 0.50, and including cash as another category gives 0.375. These are different calculation conventions and require explicit labels.

Separate securities may share industry or market exposures. A statistic calculated from allocation weights does not directly measure their return correlation.

KabuForge’s public result charts describe holding count, largest weight and HHI, available only when supporting report inputs exist. Do not treat unavailable data as zero; inspect denominators and missing-data reasons. Related explanations: maximum drawdown, the Sharpe ratio, backtesting.

Workflow and display source: public GUI research course (reviewed October 6, 2026). Fictional examples illustrate calculations and are not course results.

Public course results are research-only; historical information availability is unverified. Explaining a general definition does not establish full software support. No securities or trades are recommended. Read the disclaimer.

Explore the portfolio construction learning path

• What is rebalancing? Why target and actual weights differ