sim_SCR_data {kernscr} | R Documentation |
Data Simulation Function
Description
Data Simulation Function
Usage
sim_SCR_data(
data_size,
ncol_gene_mat,
feat_m,
feat_d,
mu_cen,
cov,
lam_m = 1/15,
lam_d = 1/20,
norm_vcov = c(1, 0.5, 0.5, 1)
)
Arguments
data_size |
an integer giving the simulated sample size |
ncol_gene_mat |
an integer giving the simulated number of genomic covariates |
feat_m |
a function that transforms the genomic features into the signal for the metastasis
process. This function should a matrix of dimensions |
feat_d |
a function that transforms the genomic features into the signal for the death
process. This function should a matrix of dimensions |
mu_cen |
mean of the exponential censoring process |
cov |
the correlation between the genomic covariates |
lam_m |
baseline hazard constant for metastasis process. Default is |
lam_d |
baseline hazard constant for death process. Default is |
norm_vcov |
vector of length 4 of correlation between errors between the two processes on
the normal scale before being complementary-log-log-transformed. Default is |
Value
a data.frame
with columns:
XR
: time to recurrence / death / censoringXD
: time to death / censoringDeltaR
: Indicator of censoring (0), recurrence (1), or death (2) for this earliest timeXR
DeltaD
: Indicator of censoring (0) or death (1)XPFS
: time to recurrence / death / censoring (=XR
)DeltaPFS
: Indicator of censoring (0) or recurrence or death, whichever came first (1)Z_1,...,Z_P
: genomic variables
Examples
feat_m_fun <- function(X){
sin(X[,1]+X[,2]^2)-1
}
feat_d_fun <- function(X){
(X[,4]-X[,5])^2/8
}
mydata <- sim_SCR_data(data_size = 400, ncol_gene_mat = 20, feat_m = feat_m_fun,
feat_d = feat_d_fun, mu_cen = 30, cov=0.5)
head(mydata)
## how many experience both events
mean(mydata[,"DeltaR"]==1 & mydata[,"DeltaD"]==1)
## how many only recur
mean(mydata[,"DeltaR"]==1 & mydata[,"DeltaD"]==0)
## how many only die
mean(mydata[,"DeltaR"]==2 & mydata[,"DeltaD"]==1)
## how many are censored
mean(mydata[,"DeltaR"]==0 & mydata[,"DeltaD"]==0)