Complex-survey GLM in one call

Builds the design with morie_survey_design() and fits morie_survey_glm(): cluster-robust, stratified design-based standard errors.

Usage

morie_survey_complex_glm(
  df,
  formula,
  weight_col,
  family = "gaussian",
  cluster_col = NULL,
  strata_col = NULL,
  nest = FALSE
)

Arguments

df

A data.frame holding the (unit-level) sample data plus any weight/strata/cluster/domain columns referenced by name.

formula

A formula (e.g. y ~ x1 + x2) for survey-weighted regression / GLM.

weight_col

Character; column name of the design weight variable in df.

family

A family object (e.g. stats::binomial()) passed to the survey-weighted GLM.

cluster_col

Character; column name of the cluster identifier in df.

strata_col

Character; column name of the stratum identifier in df.

nest

If TRUE, cluster IDs are only unique within a stratum (see morie_survey_design()).

Value

A list: coefficients (a matrix of estimates, design-based standard errors, t values and p-values on the design degrees of freedom: PSUs minus strata minus the coefficients beyond the intercept), confint, vcov (the sandwich \(A^{-1} B A^{-1}\) with \(B\) the Taylor-linearisation variance of the score totals) and fit (the weighted glm giving the estimates).

Examples

set.seed(1)
df <- data.frame(y = rnorm(40), x = rnorm(40), w = runif(40, 0.5, 2))
str(morie_survey_complex_glm(df, y ~ x, "w"), max.level = 1)
#> List of 4
#>  $ coefficients: num [1:2, 1:4] 0.11 0.123 0.135 0.14 0.813 ...
#>   ..- attr(*, "dimnames")=List of 2
#>  $ confint     : num [1:2, 1:2] -0.163 -0.16 0.382 0.407
#>   ..- attr(*, "dimnames")=List of 2
#>  $ vcov        : num [1:2, 1:2] 0.018142 0.000503 0.000503 0.019629
#>  $ fit         :List of 30
#>   ..- attr(*, "class")= chr [1:2] "glm" "lm"