模拟竞争风险数据

时间:2015-09-30 01:44:27

标签: r simulation survival-analysis

我的目标是模拟可用于测试竞争风险的数据集 模型。我只是尝试使用survsim::crisk.sim函数的简单示例 它不会导致我期望的结果。

 require(survival)
 simulated_data <- survsim::crisk.sim(n = 100,
                                      foltime = 200,
                                      dist.ev = rep("weibull", 2),
                                      anc.ev = c(0.8, 0.9),
                                      beta0.ev = c(2, 4),
                                      anc.cens = 1,
                                      beta0.cens = 5,
                                      nsit = 2)

 model <- survreg(Surv(time, status) ~ 1 + strata(cause), data = simulated_data)

 exp(model$scale)

 ## cause=1  cause=2 
 ## 4.407839 2.576357 

我希望这些数字与beta0.ev相同。什么指针 我可能会做错或其他建议如何模拟竞争风险数据。

完成:我希望模拟数据中的事件发生在每个风险不同的Weibull分布之后。我希望能够在数据中指定一个层和簇。审查可以遵循Weibull或Bernouli分布。

1 个答案:

答案 0 :(得分:1)

要恢复指定的估算值,可以将survreg与特定原因的符号一起使用。

此示例使用您的参数,但有更多患者进行更精确的估算:

set.seed(101)
stack_data <- survsim::crisk.sim(n = 2000,
                                     foltime = 200,
                                     dist.ev = rep("weibull", 2),
                                     anc.ev = c(0.8, 0.9),
                                     beta0.ev = c(2, 4),
                                     anc.cens = 1,
                                     beta0.cens = 5,
                                     nsit = 2)

m1 <- survreg(Surv(time, cause==1) ~ 1, data =stack_data, dist = "weibull")
m2 <- survreg(Surv(time, cause==2) ~ 1, data = stack_data, dist = "weibull")

m1$coefficients,原因1将接近beta0.ev

m2$coefficients,原因2将接近beta0.ev

> m1$coefficients
(Intercept) 
   1.976449 
> m2$coefficients
(Intercept) 
   3.995716 

m1$scale这将导致原因1接近anc.ev

m2$scale,原因2将接近anc.ev

> m1$scale
[1] 0.8088574
> m2$scale
[1] 0.8923334

不幸的是,这仅在统一检查或低非均匀检查(例如您的示例)中才成立

如果我们增加审查的风险,则截取不代表beta0.ev参数

set.seed(101)
stack_data <- survsim::crisk.sim(n = 2000,
                                     foltime = 200,
                                     dist.ev = rep("weibull", 2),
                                     anc.ev = c(0.8, 0.9),
                                     beta0.ev = c(2, 4),
                                     anc.cens = 1,
                                     beta0.cens = 2, #reduced from 5, increasing the hazard function for censoring rate
                                     nsit = 2)

m1 <- survreg(Surv(time, cause==1) ~ 1, data =stack_data, dist = "weibull")
m2 <- survreg(Surv(time, cause==2) ~ 1, data = stack_data, dist = "weibull")

> m1$coefficients
(Intercept) 
   1.531818 
> m2$coefficients
(Intercept) 
   3.553687 
> 
> m1$scale
[1] 0.8139497
> m2$scale
[1] 0.8910465