Extract variance-covariance matrix from a survey result object

Returns a named numeric matrix from a survey_result object. The diagonal contains per-parameter variances (se^2). Off-diagonal elements are structural zeros for most classes (except survey_corr with multiple pair rows, which uses NA_real_).

Usage

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

Arguments

object: A survey_result object from any supported get_*() function.
...: Unused.

Value

A p × p numeric matrix with rownames and colnames equal to names(coef(object)). When p > 1, emits surveycore_warning_vcov_diagonal_only. When the se column is absent, returns a p × p matrix of NA_real_ with no warning.

Details

Off-diagonal elements represent the covariance between estimates. For survey results, only the diagonal (per-estimate variances) is available from stored per-row data. Cross-estimate covariances require joint influence-function estimation from the original design. Zeros in off-diagonal positions are approximations that may not hold when groups share PSUs.

survey_corr (long-format) results with more than one pair row have NA_real_ off-diagonal elements and emit surveycore_warning_vcov_incomplete.

Examples

d <- as_survey(
  nhanes_2017,
  ids = sdmvpsu,
  weights = wtint2yr,
  strata = sdmvstra,
  nest = TRUE
)
result <- get_means(d, ridageyr, variance = "se")
vcov(result)
#>           ridageyr
#> ridageyr 0.2750013