evaluate$compare

Pairwise comparison of PRS models

Description

evaluate$compare() performs pairwise or reference-based comparison of PRS models using delta-style performance and selection metrics.

Usage

`evaluate$compare`(models = NULL, on, outcome, 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, test = c("auto", "delong", "bootstrap"), 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.
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. Compare metrics are delta-style names such as delta.auc, delta.pr.auc, delta.rmse, delta.aic, or delta.bic. Defaults depend on outcome type when NULL.
reference.model	Optional reference model used for one-vs-reference comparisons. May be supplied as a model name or 1-based model index.
test	One of “auto”, “delong”, “bootstrap”.
bootstrap	Number of bootstrap replicates for bootstrap comparison.
conf.level	Confidence level for interval estimates.
logger	Optional logger to pass and use within the function. Defaults NULL - creates a new logger

Value

A PolyGeniusEvaluation object with pairwise or reference-anchored delta rows only. Any temporary PolyGeniusData constructed from genotype input is not returned. Compare artifacts are available via slotArtifacts(), including ranking: a per-outcome, per-metric model ranking table with columns outcome, metric, rank, model.idx, model, and estimate, where rank = 1 is the top model for that outcome-metric combination. Compare diagnostics are available via slotDiagnostics(), most notably metric.flags for comparison-specific issues such as AUC test fallbacks.

See Also

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