Bayesian Semiparametric Cure Rate Model with an Unknown Threshold and Covariate Information
Posterior inference for the bayesian semiparmetric cure rate model with covariates in survival analysis.
CCuMRes( data, covs.x = names(data)[seq.int(3, ncol(data))], covs.y = names(data)[seq.int(3, ncol(data))], type.t = 3, K = 50, utao = NULL, alpha = rep(0.01, K), beta = rep(0.01, K), c.r = rep(0, K - 1), c.nu = 1, var.theta.str = 25, var.delta.str = 25, var.theta.ini = 100, var.delta.ini = 100, type.c = 4, a.eps = 0.1, b.eps = 0.1, epsilon = 1, iterations = 5000, burn.in = floor(iterations * 0.2), thinning = 3, printtime = TRUE )
data |
Double tibble. Contains failure times in the first column, status indicator in the second, and, from the third to the last column, the covariate(s). |
covs.x |
Character. Names of covariables to be part of the multiplicative part of the hazard |
covs.y |
Character. Names of covariables to determine the cure threshold por each patient. |
type.t |
Integer. 1=computes uniformly-dense intervals; 2= partition arbitrarily defined by the user with parameter utao and 3=same length intervals. |
K |
Integer. Partition length for the hazard function. |
utao |
vector. Partition specified by the user when type.t = 2. The first value of the vector has to be 0 and the last one the maximum observed time, either censored or uncensored. |
alpha |
Nonnegative entry vector. Small entries are recommended in order to specify a non-informative prior distribution. |
beta |
Nonnegative entry vector. Small entries are recommended in order to specify a non-informative prior distribution. |
c.r |
Nonnegative vector. The higher the entries, the higher the correlation of two consective intervals. |
c.nu |
Tuning parameter for the proposal distribution for c.
Only when |
var.theta.str |
Double. Variance of the proposal normal distribution for theta in the Metropolis-Hastings step. |
var.delta.str |
Double. Variance of the proposal normal distribution for delta in the Metropolis-Hastings step. |
var.theta.ini |
Double. Variance of the prior normal distribution for theta. |
var.delta.ini |
Double. Variance of the prior normal distribution for delta. from the acceptance ratio in the Metropolis-Hastings algorithm for delta*. |
type.c |
1=defines |
a.eps |
Double. Shape parameter for the prior gamma distribution of
epsilon when |
b.eps |
Double. Scale parameter for the prior gamma distribution of
epsilon when |
epsilon |
Double. Mean of the exponencial distribution assigned to
|
iterations |
Integer. Number of iterations including the |
burn.in |
Integer. Length of the burn-in period for the Markov chain. |
thinning |
Integer. Factor by which the chain will be thinned. Thinning the Markov chain reduces autocorrelation. |
printtime |
Logical. If |
Computes the Gibbs sampler with the full conditional distributions of
all model parameters (Nieto-Barajas & Yin, 2008) and arranges the resulting Markov
chain into a tibble which can be used to obtain posterior summaries. Prior
distributions for the regression coefficients Theta and Delta are assumend
independent normals with zero mean and variance var.theta.ini
,
var.delta.ini
, respectively.
It is recommended to verify chain's stationarity. This can be done by
checking each element individually. See CCuPlotDiag
.
- Nieto-Barajas, L. E., & Yin, G. (2008). Bayesian semiparametric cure rate model with an unknown threshold. Scandinavian Journal of Statistics, 35(3), 540-556. https://doi.org/10.1111/j.1467-9469.2007.00589.x
- Nieto-barajas, L. E. (2002). Discrete time Markov gamma processes and time dependent covariates in survival analysis. Statistics, 2-5.
# data(BMTKleinbook) # res <- CCuMRes(BMTKleinbook, covs.x = c("tTransplant","hodgkin","karnofsky","waiting"), # covs.y = c("tTransplant","hodgkin","karnofsky","waiting"), # type.t = 2, K = 72, length = 30, # alpha = rep(2,72), beta = rep(2,72), c.r = rep(50, 71), type.c = 2, # var.delta.str = .1, var.theta.str = 1, # var.delta.ini = 100, var.theta.ini = 100, # iterations = 100, burn.in = 10, thinning = 1)
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.