Fix term labels in check_collinearity() for brms models with 0 + Intercept syntax

[jmgirard]

For brms models fit with y ~ 0 + Intercept + x1 + x2 + ..., check_collinearity() took the term labels from labels(terms(formula)), which includes Intercept as a term. The intercept column is already dropped from the covariance matrix and the term assignment, so each VIF was shown under the name of the previous term, an Intercept row appeared, and the last term was missing. The VIF values themselves were correct.

m <- brms::brm(mpg ~ 0 + Intercept + cyl + disp + hp, data = mtcars)
check_collinearity(m)
#> before: Intercept, cyl, disp  (values belong to cyl, disp, hp)
#> after:  cyl, disp, hp

The fix removes the Intercept pseudo-term from the term labels for brmsfit models. The "Model without intercept" alert from the original report no longer fires, because insight::has_intercept() now recognizes this syntax.

A regression test fits the model above (guarded by skip_on_cran() and skip_if_not_installed("brms"), since it compiles a Stan model). It checks the term labels and compares the VIFs with the ones computed from the slope correlation matrix.

Fixes #417

[Copilot]

🟢 Approval recommended

The fix is covered by regression tests and all reviewed changes are ready for approval.

0 open findings

What changed in this PR

Fixes incorrect VIF term labels for brms models using 0 + Intercept syntax.

Changes:

• Removes the pseudo-Intercept term from labels.
• Adds guarded regression coverage for labels and VIFs.
• Updates NEWS.md and the development version.

File	Description
tests/​testthat/​test-check_collinearity.R	Adds regression coverage.
R/​check_collinearity.R	Corrects brms term-label alignment.
NEWS.md	Documents the fix.
DESCRIPTION	Bumps the development version.

🧠 Review effort: Lite

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.