Evaluation of Failure Time Surrogate Endpoints in Individual
Patient Data Meta-Analyses
Provides functions for the evaluation of
surrogate endpoints when both the surrogate and the true endpoint are failure
time variables. The approaches implemented are:
(1) the two-step approach (Burzykowski et al, 2001) with a copula model (Clayton, Plackett, Hougaard) at
the first step and either a linear regression of log-hazard ratios at the second
step (either adjusted or not for measurement error);
(2) mixed proportional hazard models estimated via mixed Poisson GLM
(Rotolo et al, 2019 ).
Evaluation of Failure Time Surrogate Endpoints in Individual Patient Data Meta-Analyses
Description
Provides functions for the evaluation of
surrogate endpoints when both the surrogate and the true endpoint are failure
time variables. The approaches implemented are: (1) the two-step approach
(Burzykowski et al, 2001) DOI:10.1111/1467-9876.00244
with a copula model (Clayton, Plackett, Hougaard) at
the first step and either a linear regression of log-hazard ratios at the second
step (either adjusted or not for measurement error); (2) mixed proportional
hazard models estimated via mixed Poisson GLM.
News
Changes in version 1.1.23 (2017-07-13)
new function simData.gh() to generate data from Gumbel-Hougaard copula
Changes in version 1.1.15 (2017-04-25)
fixed minor issue in prediction function for adjusted copula models
Changes in version 1.1.14 (2017-04-06)
fixed issue in prediction intervals for mixed Poisson models
Changes in version 1.1.13 (2017-03-28)
fixed issue in prediction intervals for adjusted copula models
Changes in version 1.1.12 (2017-03-02)
loocv() now also returns the values of kTau and R2
estimated in each (N-1) fold
Changes in version 1.1.10 (2017-02-27)
fixed issues with loocv when few trials (added controls)
added data 'gastadj'
added twoStage parameter to surrosurv for Shih and Louis (1995) approach
to copula estimation
new function ste() to compute the surrogate threshold effect
plot.sussosurv() can now show prediction intervals and the STE
Changes in version 0.0.11 (2016-08-12)
predict and plot for class sussosurv
Changes in version 0.0.10 (2016-08-01)
kkt2 convergence criteria corrected from positive determinant
to positive min eigenvalue
Changes in version 0.0.9 (2016-07-28)
bugfix in Poisson method, which did not return results
because of mispelled model name
Changes in version 0.0.7 (2016-07-25)
the Kendall's tau for copulas is now initialized
using the SurvCorr package (much faster)
Changes in version 0.0.6 (2016-07-22)
added the function convals(), giving the values of the max abs gradient
and the min eigenvalue of the variance-covariance matrix of the random treatment effects
the function convergence() uses explicit computation provided by covals(),
instead of using the function optimx in the package optimx