Convert survival parameters to SAS/STEM parametric forms

convSTEM(x = NULL, samples = NULL, use = "everything")

Arguments

x

An object created by calling runPSM

samples

An object created by calling boot with bootPSM

use

an optional character string giving a method for computing covariances in the presence of missing values. See cov for details. Option "complete.obs" maybe needed when some bootstrap samples do not converge to estimate covariance only using those that do.

This function primarily exists for backward compatibility with older excel models where parametric extrapolation was performed with SAS and alternative parametric forms were used for distributions. As such only a subset of models are supported. One or both of x and samples must be specified and affect what is returned. For more details please see the vignette vignette("STEM_compatability", package = "flexsurvPlus")

Possible distributions include

  • Exponential ('exp')

  • Weibull ('weibull')

  • Gompertz ('gompertz')

  • Log-normal ('lnorm')

  • Log-logistic ('llogis')

  • Generalized gamma ('gengamma')

  • Gamma ('gamma')

Value

a list containing 4 data frames

  • stem_param Converted parameter estimates

  • stem_cov Converted covariance matrix (if samples provided)

  • stem_modsum Converted model summary (if x provided)

  • stem_boot Converted bootstrap samples (if samples provided)