Skip to contents

Extract quantiles of the potential impact fraction or population attributable fraction bootstrap simulations

quantile_bootstrap.Rd

Method for obtaining the variance-covariance matrix of bootstrap simulations of a pif_class object (i.e. the point-estimate of the potential impact fraction or population attributable fraction via the bootstrap).

Usage

quantile_bootstrap(x, prob, ...)

Arguments

x

A pif_class object

prob

numeric vector of probabilities with values between 0 and 1

...

additional arguments to pass to stats::quantile

Examples

#Example 1
data(ensanut)
options(survey.lonely.psu = "adjust")
design <- survey::svydesign(data = ensanut, ids = ~1, weights = ~weight, strata = ~strata)
rr <- function(X, theta) {
  exp(-2 +
    theta[1] * X[, "age"] + theta[2] * X[, "systolic_blood_pressure"] / 100)
}
cft <- function(X) {
  X[, "systolic_blood_pressure"] <- X[, "systolic_blood_pressure"] - 5
  return(X)
}
pifsim <- pif(design,
  theta = log(c(1.05, 1.38)), rr, cft,
  additional_theta_arguments = c(0.01, 0.03), n_bootstrap_samples = 10,
)
quantile_bootstrap(pifsim, c(0.1, 0.3, 0.5))
#>     counterfactual   relative_risk potential_impact_fraction
#> 1 Counterfactual_1 Relative_Risk_1               0.006058022
#> 2 Counterfactual_1 Relative_Risk_1               0.013733630
#> 3 Counterfactual_1 Relative_Risk_1               0.017201760
#>   average_relative_risk average_counterfactual quantile
#> 1           0.002041747            0.001998326      0.1
#> 2           0.277577252            0.270856565      0.3
#> 3           2.214570718            2.183890090      0.5