Calibrate logistic coefficients to prevalence and treatment contrast

Usage

calibrate_logistic(
  data,
  coefficients,
  prevalence,
  risk_ratio = NULL,
  risk_difference = NULL,
  treatment = "treatment",
  treatment_prevalence = 0.5,
  auc = NULL,
  tolerance = 1e-08
)

Arguments

data: Covariate data.
coefficients: Named covariate coefficients.
prevalence: Target population prevalence.
risk_ratio: Optional treated/control risk ratio.
risk_difference: Optional treated-control risk difference.
treatment: Name for the optional treatment coefficient.
treatment_prevalence: Treatment prevalence used for calibration.
auc: Optional target population area under the ROC curve. Covariate coefficients are proportionally scaled to reach this target.
tolerance: Numerical root tolerance.

Value

A base data.frame with one row per calibrated coefficient and columns term, coefficient, calibration and coefficient_scale. Coefficients are log-odds on the logit scale. The first row is "(Intercept)", followed by the treatment row when risk_ratio or risk_difference was supplied, then one row per name of coefficients. calibration names the target that was solved for – "prevalence", "auc", "risk_ratio" or "risk_difference" – and coefficient_scale is the factor the supplied covariate coefficients were multiplied by, which is 1 unless auc was given.

Examples

set.seed(1)
covariates <- data.frame(x1 = stats::rnorm(200), x2 = stats::rnorm(200))

calibrate_logistic(
  covariates, coefficients = c(x1 = 0.5, x2 = -0.3), prevalence = 0.2
)
#>          term coefficient calibration coefficient_scale
#> 1 (Intercept)    -1.48188  prevalence                 1
#> 2          x1     0.50000  prevalence                 1
#> 3          x2    -0.30000  prevalence                 1

# Scale the coefficients to hit a target discrimination instead.
calibrate_logistic(
  covariates, coefficients = c(x1 = 0.5, x2 = -0.3),
  prevalence = 0.2, auc = 0.75
)
#>          term coefficient calibration coefficient_scale
#> 1 (Intercept)  -1.6789042         auc          1.843123
#> 2          x1   0.9215615         auc          1.843123
#> 3          x2  -0.5529369         auc          1.843123