Uses a Metropolis Hastings scheme on the proportional hazards model to draw sample from posterior. Uses a matched curvature Student's t candidate generating distribution with 4 degrees of freedom to give heavy tails.
BayesCPH(
y,
t,
x,
steps = 1000,
priorMean = NULL,
priorVar = NULL,
mleMean = NULL,
mleVar,
startValue = NULL,
randomSeed = NULL,
plots = FALSE
)the Poisson censored response vector. It has value 0 when the variable is censored and 1 when it is not censored.
time
matrix of covariates
the number of steps to use in the Metropolis-Hastings updating
the mean of the prior
the variance of the prior
the mean of the matched curvature likelihood
the covariance matrix of the matched curvature likelihood
a vector of starting values for all of the regression coefficients including the intercept
a random seed to use for different chains
Plot the time series and auto correlation functions for each of the model coefficients
A list containing the following components:
a data frame containing the sample of the model coefficients from the posterior distribution
the mean of the matched curvature likelihood. This is useful if you've used a training set to estimate the value and wish to use it with another data set
the covariance matrix of the matched curvature likelihood. See mleMean for why you'd want this