Derive a scoring rule for PDD — derive_scoring

Given a prediction linear logistic model and its ROC analysis results, computes a scoring rule for deciding a patient suffers probable PDD given screening criteria.

Usage

derive_scoring_rule(pt, coefs, inverse = TRUE, ...)

Arguments

pt: The threshold on probability scale. E.g., derived from run_roc.
coefs: Model coefficients as computed by extract_coefficients().
inverse: Should pt be inversed from probability to logit scale?
...: Unused. There to allow for compatibility with run_scoring_rule_pipeline.

Value

List containing:

score coefficients
cutoff
equation (character)
integer scoring rule