evaluate$benchmark

Unified benchmark evaluation across PRS analyses

Description

evaluate$benchmark() orchestrates multiple evaluation analyses in one call: discrimination, calibration, pairwise comparison, similarity, risk strata, and incremental value (when baseline covariates are provided).

Usage

`evaluate$benchmark`(models = NULL, on, outcome, baseline.covariates = NULL, type = c("auto", "binary", "continuous", "survival"), time = NULL, event = NULL, obs = NULL, scores.layer = X, score.mode = c("compute.if.missing", "require", "recompute"), score.args = list(), metrics = NULL, reference.model = NULL, compare.test = c("auto", "delong", "bootstrap"), similarity.source = c("scores", "variants"), quantiles = c(0.2, 0.8), bootstrap = 2000, conf.level = 0.95)

Arguments

models	Optional model specification.
on	Evaluation context (PolyGeniusData or genotype input). When genotype input is supplied, PolyGenius internally materializes a temporary PolyGeniusData object to resolve and evaluate scores.
outcome	Outcome definition. When on is PolyGeniusData: unquoted expression resolved on observations. When on is genotype input: vector of length n_obs, list of vectors (each length n_obs), or table with one or more columns and nrow == n_obs.
baseline.covariates	Optional unquoted baseline covariates expression.
type	Outcome type (“auto”, “binary”, “continuous”, “survival”).
time	Unquoted time-to-event expression (required for survival).
event	Unquoted event-indicator expression (required for survival).
obs	Optional unquoted observation subset expression.
scores.layer	Score layer to read/use (symbol or single string).
score.mode	Score resolution mode. If on is a PolyGeniusData object, computed scores are written into that object. If on is genotype input, computed scores exist only in the temporary internal evaluation data object and are not returned.
score.args	Named list passed to compute$scores(…) when needed.
metrics	Optional metric subset across analyses. When NULL, defaults are used for each analysis.
reference.model	Optional reference model passed to evaluate$compare() when pairwise comparison results should be anchored to one model.
compare.test	Pairwise-comparison test (“auto”, “delong”, “bootstrap”).
similarity.source	Similarity source passed to evaluate$similarity(). Defaults to “scores”.
quantiles	Quantiles used by risk-strata analysis.
bootstrap	Number of bootstrap replicates for supported analyses.
conf.level	Confidence level for interval estimates.

Value

A PolyGeniusEvaluation object with merged benchmark result rows. Any temporary PolyGeniusData constructed from genotype input is not returned. The returned object also carries the merged artifacts, diagnostics, and benchmark log assembled from the component evaluations.

See Also

Other evaluate: evaluate.calibration(), evaluate.compare(), evaluate.discrimination(), evaluate.incremental(), evaluate.risk.strata(), evaluate.similarity()