This function computes posterior mode estimates of the parameters of a flexible Cox model
with structured additive predictors using a Newton-Raphson algorithm. Integrals are solved
numerically. Moreover, optimum smoothing variances are computed using a stepwise optimization,
see also the details section of function
cox_mode(x, y, start, weights, offset, criterion = c("AICc", "BIC", "AIC"), nu = 0.1, update.nu = TRUE, eps = .Machine$double.eps^0.25, maxit = 400, verbose = TRUE, digits = 4, ...)
The model response, as returned from function
A named numeric vector containing possible starting values, the names are based on
Prior weights on the data, as returned from function
Can be used to supply model offsets for use in fitting,
returned from function
Set the information criterion that should be used, e.g., for smoothing
variance selection. Options are the corrected AIC
Calibrates the step length of parameter updates of one Newton-Raphson update.
Should the updating step length be optimized in each iteration of the backfitting algorithm.
The relative convergence tolerance of the backfitting algorithm.
The maximum number of iterations for the backfitting algorithm
Print information during runtime of the algorithm.
Set the digits for printing when
Currently not used.
A list containing the following objects:
A named list of the fitted values of the modeled parameters of the selected distribution.
The estimated set regression coefficients and smoothing variances.
The equivalent degrees of freedom used to fit the model.
The value of the log-likelihood.
The value of the log-posterior.
The Hessian matrix evaluated at the posterior mode.
Logical, indicating convergence of the backfitting algorithm.
The runtime of the algorithm.
Umlauf N, Klein N, Zeileis A (2016). Bayesian Additive Models for Location Scale and Shape (and Beyond). (to appear)
## Please see the examples of function cox_mcmc()!