evaluate$calibration

Calibration metrics for PRS models

Description

evaluate$calibration() assesses agreement between predictions and observed outcomes for one or more PRS models.

Usage

evaluate$calibration(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, n.groups = 10, conf.level = 0.95)

Arguments

models	Optional model specification (see evaluate$discrimination()).
on	Evaluation context (PolyGeniusData or genotype input). When genotype input is supplied, PolyGenius internally materializes a temporary PolyGeniusData object to resolve and evaluate scores.
outcome	Outcome definition. When on is PolyGeniusData: unquoted expression resolved on observations. When on is genotype input: vector of length n_obs, list of vectors (each length n_obs), or table with one or more columns and nrow == n_obs.
type	Outcome type (“auto”, “binary”, “continuous”, “survival”).
time	Unquoted time-to-event expression (required for survival).
event	Unquoted event-indicator expression (required for survival).
obs	Optional unquoted observation subset expression.
scores.layer	Score layer to read/use (symbol or single string).
score.mode	Score resolution mode. If on is a PolyGeniusData object, computed scores are written into that object. If on is genotype input, computed scores exist only in the temporary internal evaluation data object and are not returned.
score.args	Named list passed to compute$scores(…) when needed.
metrics	Optional metric subset; defaults by outcome type when NULL.
n.groups	Number of groups for Hosmer-Lemeshow style grouping (binary outcomes).
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 only. Any temporary PolyGeniusData constructed from genotype input is not returned. Calibration artifacts are available via slotArtifacts():

• calibration for all outcome types: one row per evaluated observation and model with observed and predicted values. For survival outcomes, predicted is the fitted linear predictor and observed is the event indicator.

• deciles for binary outcomes: one row per Hosmer-Lemeshow grouping bin with group, observed.events, expected.events, n, observed.rate, and expected.rate.

See Also

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