I am going through R's function indeptCoxph() in the spBayesSurv package which fits a bayesian Cox model. I am confused by some of the input parameters to this functions. (I also had some questions about the R code which I have posted separately on Stack Overflow: Stuck with package example code in R - simulating data to fit a model).

What is the role of the "prediction" input parameter? Should it not only contain the predictor covariates? In the R example, the authors have included a vector "s" which was used to initially simulate the survival times data in their example as well as the predictors. I'm not sure what this "s" is.

Given that my data is just a set of survival times between 0 and 100, along with censored (yes/no) information, how would I use this function and how should I handle the input "s"?

(I have also posted on SO, but posting here too since I would like to understand the theory behind this model ).

I am going through R's function indeptCoxph() in the spBayesSurv package which fits a bayesian Cox model. I am confused by some of the input parameters to this functions. (I also had some questions about the R code which I have posted separately on Stack Overflow: Stuck with package example code in R - simulating data to fit a model).

What is the role of the "prediction" input parameter? Should it not only contain the predictor covariates? In the R example, the authors have included a vector "s" which was used to initially simulate the survival times data in their example as well as the predictors. I'm not sure what this "s" is.

Given that my data is just a set of survival times between 0 and 100, along with censored (yes/no) information, how would I use this function and how should I handle the input "s"?

(I have also posted on SO, but posting here too since I would like to understand the theory behind this model ).

I am going through R's function indeptCoxph in the spBayesSurv package which fits a bayesian Cox model. I am confused by some of the input parameters to this functions. ( I also had some questions about the R code which I have posted separately here: Stuck with package example code in R - simulating data to fit a model ).

What is the role of the "prediction" input parameter? Should it not only contain the predictor covariates? In the R example, the authors have included a vector "s" which was used to initially simulate the survival times data in their example as well as the predictors. I'm not sure what this "s" is.

Given that my data is just a set of survival times between 0 and 100, along with censored (yes/no) information, how would I use this function and how should I handle the input "s"?

(I have also posted on SO, but posting here too since I would like to understand the theory behind this model ).

I am going through R's function indeptCoxph() in the spBayesSurv package which fits a bayesian Cox model. I am confused by some of the input parameters to this functions. (I also had some questions about the R code which I have posted separately on Stack Overflow: Stuck with package example code in R - simulating data to fit a model).

What is the role of the "prediction" input parameter? Should it not only contain the predictor covariates? In the R example, the authors have included a vector "s" which was used to initially simulate the survival times data in their example as well as the predictors. I'm not sure what this "s" is.

Given that my data is just a set of survival times between 0 and 100, along with censored (yes/no) information, how would I use this function and how should I handle the input "s"?

(I have also posted on SO, but posting here too since I would like to understand the theory behind this model ).