R语言代写如何在生存分析与Cox回归中计算IDI，NRI指标

 原文链接：http://tecdat.cn/?p=6095

读取样本数据

 D$status=as.numeric(D$status==2)
D=D[!is.na(apply(D,1,mean)),] ; dim(D)

## [1] 416   7

查询部分数据（结果和预测因子）

head(D)

##   time status      age albumin edema protime bili
## 1  400      1 58.76523    2.60   1.0    12.2 14.5
## 2 4500      0 56.44627    4.14   0.0    10.6  1.1
## 3 1012      1 70.07255    3.48   0.5    12.0  1.4
## 4 1925      1 54.74059    2.54   0.5    10.3  1.8
## 5 1504      0 38.10541    3.53   0.0    10.9  3.4
## 6 2503      1 66.25873    3.98   0.0    11.0  0.8

模型0和模型1的结果数据和预测变量集

outcome=D[,c(1,2)]
covs1<-as.matrix(D[,c(-1,-2)])
covs0<-as.matrix(D[,c(-1,-2, -7)])

head(outcome)

##   time status
## 1  400      1
## 2 4500      0
## 3 1012      1
## 4 1925      1
## 5 1504      0
## 6 2503      1

head(covs0)

##        age albumin edema protime
## 1 58.76523    2.60   1.0    12.2
## 2 56.44627    4.14   0.0    10.6
## 3 70.07255    3.48   0.5    12.0
## 4 54.74059    2.54   0.5    10.3
## 5 38.10541    3.53   0.0    10.9
## 6 66.25873    3.98   0.0    11.0

head(covs1)

##        age albumin edema protime bili
## 1 58.76523    2.60   1.0    12.2 14.5
## 2 56.44627    4.14   0.0    10.6  1.1
## 3 70.07255    3.48   0.5    12.0  1.4
## 4 54.74059    2.54   0.5    10.3  1.8
## 5 38.10541    3.53   0.0    10.9  3.4
## 6 66.25873    3.98   0.0    11.0  0.8

推理  

<span style="color:#333333"><span style="color:#333333"><code><span style="color:#000000">t0</span><span style="color:#687687">=</span><span style="color:#009999">365</span><span style="color:#687687">*</span><span style="color:#009999">5</span>
<span style="color:#000000">x</span><span style="color:#687687"><-</span><span style="color:#000000">IDI </span><span style="color:#687687">(</span><span style="color:#000000">outcome</span>, <span style="color:#000000">covs0</span>, <span style="color:#000000">covs1</span>, <span style="color:#000000">t0</span>, <span style="color:#000000">npert</span><span style="color:#687687">=</span><span style="color:#009999">200</span><span style="color:#687687">)</span> ;</code></span></span>

输出 

##     Est. Lower Upper p-value
## M1 0.090 0.052 0.119       0
## M2 0.457 0.340 0.566       0
## M3 0.041 0.025 0.062       0

M1表示IDI

M2表示NRI

M3表示中位数差异

图形演示

​

如果您有任何疑问，请在下面发表评论。