Skip to contents

Native E-values for the EValue dispatch family (evalues.OLS, evalues.RR, evalues.OR, evalues.HR, evalues.MD); OLS and MD use the \(\exp(0.91 d \pm 1.78 se)\) interval. Pairs with e_value_rr / e_value_or / e_value_hr / e_value_d.

Usage

morie_sensitivity_evalue(
  estimate,
  se = NULL,
  sd = NULL,
  type = c("OLS", "RR", "OR", "HR", "MD"),
  rare = TRUE,
  true = NULL,
  ci_lower = NULL,
  ci_upper = NULL,
  ...
)

Arguments

estimate

Observed effect on the requested scale.

se

Standard error of estimate (used by "OLS" and "MD").

sd

Outcome standard deviation (required by "OLS").

type

One of "OLS" (default), "RR", "OR", "HR", "MD".

rare

Logical; only relevant for "OR" and "HR" (passed to EValue). Default TRUE (rare-outcome approximation).

true

Reference value on the appropriate scale (default 0 for OLS / MD, 1 for ratio scales).

ci_lower, ci_upper

Optional 95% CI on the same scale as estimate.

...

Unused.

Value

A list of class morie_sensitivity_evalue with estimate, e_value_point, e_value_ci, type, method, and raw (the point and CI E-values).

References

VanderWeele, T. J., & Ding, P. (2017). Sensitivity analysis in observational research: introducing the E-value. Annals of Internal Medicine, 167(4), 268–274.

Examples

str(morie_sensitivity_evalue(1.8, type = "RR", ci_lower = 1.2,
                             ci_upper = 2.7), max.level = 1)
#> List of 6
#>  $ estimate     : num 1.8
#>  $ e_value_point: num 3
#>  $ e_value_ci   : num 1.69
#>  $ type         : chr "RR"
#>  $ method       : chr "evalues.RR (native)"
#>  $ raw          :List of 2
#>  - attr(*, "class")= chr [1:2] "morie_sensitivity_evalue" "list"