From 17af45b984d348505527140470c7fff204748ea6 Mon Sep 17 00:00:00 2001 From: Johannes Ranke Date: Fri, 12 Sep 2025 22:16:13 +0200 Subject: Facilitate centering covariates The motivation for this is to obtain confidence intervals for covariate dependent parameters that are valid for a median/mean value of the covariate. Implementation by adding an argument to the 'saem' function, and adapting the relevant functions working with 'saem' objects. Vignettes, the template and tests were updated to use the new functionality. --- man/saem.Rd | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'man/saem.Rd') diff --git a/man/saem.Rd b/man/saem.Rd index fd19fe59..5d42f1b1 100644 --- a/man/saem.Rd +++ b/man/saem.Rd @@ -22,6 +22,7 @@ saem(object, ...) omega.init = "auto", covariates = NULL, covariate_models = NULL, + center_covariates = c(NA, "median", "mean"), no_random_effect = NULL, error.init = c(1, 1), nbiter.saemix = c(300, 100), @@ -101,6 +102,10 @@ initialisation of \link[saemix:saemixModel]{saemix::saemixModel} is used for ome variable, i.e. of the type 'parameter ~ covariate'. Covariates must be available in the 'covariates' data frame.} +\item{center_covariates}{Either NA, for no centering, or your +preferred function for calculating the center, currently either +median or mean.} + \item{no_random_effect}{Character vector of degradation parameters for which there should be no variability over the groups. Only used if the covariance model is not explicitly specified.} -- cgit v1.2.3