imputeYn (version 1.3)

data: Generating Survival Data from Log-normal AFT Model

Description

This gives the survival data generated from log-normal AFT model.

Usage

data(n, p, r, b1, sig, Cper)

Arguments

n

sample size.

p

the number of covariates. For the AFT model each covariate is generated from Uniform(0, 1) distribution

r

correlation between the covariates, r is set to 0 for no correlation.

b1

the vector of coefficients.

sig

this maintains noise ratio, 1 for no noise.

Cper

takes specific value for generating specific censoring percentage, e.g., -0.2 for 30 censoring percentage, 0.0 for 50 censoring percentage and 0.2 for 70 percentages.

Value

y

logarithmic of survival time

x

matrix of covariates of order n by p

delta

status; 1 for uncensored, o for censored

Pper

censoring percentage

Details

Generate survival data from a log-normal AFT model (Y = alpha + X (beta) + error; Y=log(T)) where error is N(0,1). The last largest datum is generated always as censored otherwise censorship is random with censoring time generated from Uniform (c, 2c) for a suitable c.

References

Khan and Shaw. (2013a). On Dealing with Censored Largest Observations under Weighted Least Squares. CRiSM working paper, Department of Statistics, University of Warwick, UK, No. 13-07. Also available in http://arxiv.org/abs/1312.2533.

Examples

#Dataset with zero correlation between the covariates and the medium censoring level 
#(50 percent) 
data1<-data(n=100, p=2, r=0, b1=c(2,4), sig=1, Cper=0)
data1

#Dataset with moderate correlation between the covariates and the higher censoring level 
#(70 percent) 
data.r<-data(n=100, p=2, r=0.5, b1=c(2,4), sig=1, Cper=0.2)
data.r