get_covbeta function

Compute covariance of fixed effect parameters as a function of variance parameters of a linear mixed model

Compute covariance of fixed effect parameters as a function of variance parameters of a linear mixed model

At the optimum the covariance is available as vcov(lmer-model). This function computes cov(beta) at non (RE)ML estimates of varpar.

get_covbeta(varpar, devfun)

Arguments

  • varpar: variance parameters; varpar = c(theta, sigma).
  • devfun: deviance function as a function of theta only.

Returns

The covariances matrix of the fixed effects at supplied varpar values.

Author(s)

Rune Haubo B. Christensen. Adapted to pbkrtest by Søren Højsgaard.