generate_stan_data_2: Bayesian proportional hazards regression

In csetraynor/rstanhaz: Bayesian Survival Models (rstanhaz)

Description

Bayesian inference for proportional hazards regression models. The user can specify a variety of standard parametric distributions for the baseline hazard, or a Royston-Parmar flexible parametric model.

Usage

generate_stan_data_2(formula1, formula2, data, basehaz = "fpm",
  timescale = "log", df = 5L, degree = 3L, iknots = NULL,
  bknots = NULL, prior_1 = normal(), prior_2 = hs(),
  prior_intercept = normal(), add_prior = NULL, prior_aux = list(),
  prior_PD = FALSE, algorithm = c("sampling", "meanfield", "fullrank"),
  adapt_delta = 0.95, max_treedepth = 11L, init = "random", cores = 1L,
  out_data = FALSE, ...)

Examples

pbc2 <- survival::pbc
pbc2 <- pbc2[!is.na(pbc2$trt),]
pbc2$status <- as.integer(pbc2$status > 0)
m1 <- stan_surv(survival::Surv(time, status) ~ trt, data = pbc2)

df <- flexsurv::bc
m2 <- stan_surv(survival::Surv(rectime, censrec) ~ group,
                data = df, cores = 1, chains = 1, iter = 2000,
                basehaz = "fpm", iknots = c(6.594869,  7.285963 ),
                degree = 2, prior_aux = normal(0, 2, autoscale = F))

csetraynor/rstanhaz documentation built on May 9, 2019, 8:14 a.m.