Efficient Estimation of Discrete-Time Multivariate Frailty Model
Using Exact Likelihood Function for Grouped Survival Data
Description
The core of this 'Rcpp' based package is several functions to
compute the baseline hazard, effect parameter, and frailty variance for the
discrete-time shared frailty model with random effects. The core functions
include two processes: (1) evaluate the multiple variable integration to
compute the exact proportional hazards model based likelihood and (2) estimate
desired parameters using maximum likelihood estimation. The integration is
evaluated by 'Cuhre' function from 'Cuba' library (Hahn, T., Cuba-a library
for multidimensional numerical integration, Comput. Phys. Commun. 168, 2005,
78-95 ), and the source files of 'Cuhre' function are included in this package.
Maximization process is carried out using the Brent's algorithm, and the 'C+
+' code file is from John Burkardt and John Denker (Brent, R.,Algorithms for
Minimization without Derivatives, Dover, 2002, ISBN 0-486-41998-3).