Nesting vs Grouping (Crossed Structures)

1. Why This Matters

Many datasets in econometrics, panel data, and hierarchical modeling involve structured observations
individuals within groups, or observations classified by multiple dimensions.

Understanding whether these structures are nested or crossed determines:

  • How we model random effects / fixed effects
  • How we cluster standard errors
  • How we interpret within- and between-group variation

2. Nested Data Structures

Definition A nested structure means smaller units are fully contained within larger units.
Each lower-level unit belongs to exactly one higher-level unit.

\([ \text{Level 3: Country} ; ⊃ ; \text{Level 2: Region} ; ⊃ ; \text{Level 1: City} ]\)

Key Properties

Feature Explanation
Belonging One-to-one hierarchy
Independence Observations within the same group are correlated
Model implication Use hierarchical or multilevel models with random intercepts/slopes
Typical notation a/b/c (e.g., country/region/city)

Economic Examples

Example Interpretation
Students within Schools Each student belongs to one school
Cities within Regions within Countries Strict geographic hierarchy
Years within Firms Time nested within individual firms (panel data)

Model Form

\(Y_{ij} = \beta_0 + u_j + \beta_1 X_{ij} + e_{ij}\)

where \(u_j\) = group-level (nested) random effect.


3. Grouped or Crossed Data Structures

Definition

A grouped (crossed) structure means units are classified along multiple independent dimensions.
Each observation can belong to several groups simultaneously.

$ $

Key Properties

Feature Explanation
Belonging Many-to-many relationship
Independence Observations share multiple group memberships
Model implication Use crossed random effects (or two-way fixed effects)
Typical notation a * b (e.g., city * brand)

Economic Examples

Example Interpretation
Teachers × Students Each teacher teaches many students; each student has multiple teachers
Firms × Industries Firms operate across industries; industries have many firms
Cities × Brands Each city sells many brands; each brand sold in many cities

Model Form

\(Y_{ij} = \beta_0 + u_i^{(city)} + v_j^{(brand)} + e_{ij}\)

where \(u_i\), \(v_j\) = crossed random effects.


4. Mixed (Nested + Crossed) Structures

Sometimes, you have both:

\((\text{country/region/city}) * (\text{brand/product})\)

Interpretation:

  • Cities are nested within regions and countries.
  • Brands/products are nested within brand lines.
  • The two hierarchies cross each other → mixed structure.

Example:

Country Region City Brand Product Sales
USA California Los Angeles Nike Air Max 300
USA California Los Angeles Apple iPhone 500
UK England London Nike Air Max 220

Model Form:

$Y_{ijkm} = 0 + u{country_j} + u_{region_{k(j)}} + u_{city_{m(k,j)}} + v_{brand_b} + v_{product_{p(b)}} + e_{ijkm} $

where \(u\) = nested random effects; \(v\) = crossed random effects.


5. Summary Table

Concept Relationship Example Model Type Typical Notation
Nesting One unit belongs exclusively to another Students within schools Multilevel / hierarchical a/b/c
Grouping (Crossed) Units classified by multiple factors City × Brand sales Crossed random / two-way FE a * b
Mixed Nested hierarchies crossed with another (Country/Region/City) × (Brand/Product) Mixed-effects (a/b) * (c/d)

##6. Key Takeaways

  • Nesting = contained within (hierarchy).
    → one-to-one belonging (tree structure).
  • Grouping = classified by (intersection).
    → many-to-many relationship (grid structure).
  • Mixed structures combine both.
    → e.g., “Cities within Countries × Brands across Markets.”

*Econometric translation:**

Data Type Model Framework
Time within Firm Fixed or random effects (nested)
Individuals across Firms × Industries Two-way fixed effects (crossed)
Schools within Districts × Teachers across Classes Mixed / Crossed multilevel

Summary

Nesting forms hierarchies; grouping forms grids.
Econometric models must reflect which structure your data actually follow.