Variance-covariance matrix for a beezdemand_tmb fit

Returns the fixed-effect VCOV from the TMB sdreport, i.e., the inverse of the negative Hessian at the MLE after Laplace-marginalizing the random effects. Row/column names follow the optimizer's internal parameterization (matching names(coef(object, type = "internal"))).

Usage

# S3 method for class 'beezdemand_tmb'
vcov(object, ...)

Arguments

object

A beezdemand_tmb object.

...

Unused.

Value

Numeric symmetric matrix of dimension p x p.

See also

coef.beezdemand_tmb(), confint.beezdemand_tmb().

Examples

# \donttest{
data(apt)
fit <- fit_demand_tmb(apt, equation = "exponential", verbose = 0)
#>   equation='exponential': Dropped 14 zero-consumption observations (146 remaining).
V <- vcov(fit)
isSymmetric(V)
#> [1] TRUE
# }