checkConv: Extended Convergence Checking

In lme4: Linear Mixed-Effects Models using 'Eigen' and S4

Extended Convergence Checking

Description

Primarily internal code for checking optimization convergence, see convergence for a more detailed discussion.

Usage

checkConv(derivs, coefs, ctrl, lbound, ubound,
          debug = FALSE, nobs = NULL, ndim = NULL)

Arguments

derivs	typically the "derivs" attribute of optimizeLmer(); with "gradients" and possibly "Hessian" component
coefs	current coefficient estimates
ctrl	list of lists, each with action character strings specifying what should happen when a check triggers, and tol numerical tolerances, as is the result of lmerControl()$checkConv.
lbound, ubound	vectors of lower and upper bounds for random-effects parameters only (length is taken to determine number of RE parameters)
debug	enable debugging output, useful if some checks are on "ignore", but would "trigger"
nobs	number of observations (needed to determine whether to do convergence checks for large data)
ndim	number of dimensions for the variance-covariance matrix of random effects (needed to determine whether to do the gradient/hessian checks for large nonlinear optimization problems)

Value

A result list containing

code	The return code for the check
messages	A character vector of warnings and messages generated by the check

See Also

convergence

lme4 documentation built on March 6, 2026, 1:07 a.m.