timeROC package (v0.4) is being archived
on CRAN. Its required functions have been integrated directly into
CalibrationCurves (R/timeROC_archived.R) with proper attribution. timeROC
has been removed from Imports; pec has been added instead (required for
ipcw()).valProbCluster(): cl.level parameter was ignored (Issue #22): The
confidence level was hardcoded to 0.95 throughout the clustering pipeline.
cl.level is now correctly passed to CGC(), MAC2(), and MIXC() and
propagated to metaprop(), metagen(), and rma.mv() calls. Hardcoded
"95%" plot labels have been replaced with dynamic labels via the new
ci_pi_labels() helper function.
valProbSurvival(): crash near max follow-up (Issue #24): Uno's
time-dependent AUC was evaluated at max(fit$y) - 0.01 instead of the
user-specified timeHorizon, causing "incorrect number of dimensions" errors
when the risk set was depleted. Fixed to use times = timeHorizon.
New "default" approach in valProbCluster(): Combines MAC2 (splines)
for the overall calibration curve, confidence intervals, and prediction
intervals, with MIXC for cluster-specific curves. This is now the default
when approach is not specified. The returned object contains both the MAC2
overall results (results$overall) and the MIXC cluster results
(results$clusters).
Unified the legend title to "Heterogeneity" across all valProbCluster()
approaches.
Updated plot font to sans-serif and increased base size to 11 for improved readability across MIXC, MAC2, and CGC approaches.
cl.level in valProbCluster().ci_pi_labels() for formatting confidence/prediction
interval labels dynamically."default" approach to the package vignette.valProbCluster() for calibration of clustered data with three
approaches: MIXC, MAC2, and CGC.valProbSurvival() for calibration of survival/time-to-event data.genCalCurve() for generalized calibration across the exponential
family.Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.