I would like to implement the time-varying coefficients model (cf. Fan and Zhang, 1999) for a Cox proportional hazards model, as proposed by Cai and Sun (2003), and studied by Tian, Zucker and Wei (2005).
This is a somewhat under-appreciated approach to a basic problem. There are commonly used related methods, such as in base R survival that make parametric assumptions on the form of the time-varying coefficients. By contrast I am looking for a local smoothing approach to the estimating equation.
I would like to know if there is a publicly available implementation, particularly for R. If not I would like advice to implement the method. Do I need to start from scratch or can I simply start from the output of standard Cox software such as coxph which gives the gradient, hessian, and allows subject specific weights?