循环计算逻辑回归模型

统计计算模型的P值或者截距等变量是常有的事情，一般在变量不多的情况下，多数一个一个的手动计算，但不免有时原始数据的改动，导致一系列的变动，很不好受。
这时R语言的作用就很明显，只需要run一下，就可以重新处理一遍数据。
同时R语言中的for循环或者apply族循环，也是解放双手的一大利器。
今天举个小栗子，说说循环这件事。

1 数据格式

• 136行27列，
• 27个变量，第一个变量与其余26个变量都需要建立一一联系

> str(data)
'data.frame':	136 obs. of  27 variables:
 $ recurrence       : int  1 0 1 0 1 1 1 0 0 0 ...
 $ Etiology         : Factor w/ 5 levels "1","2","3","4",..: 1 1 4 2 3 1 4 1 1 1 ...
 $ Child.pugh       : Factor w/ 3 levels "1","2","3": 1 2 1 1 1 2 2 2 1 2 ...
 $ Ascites          : Factor w/ 4 levels "0","1","2","3": 2 3 3 1 2 3 3 4 1 3 ...
 $ ALBI.stage       : Factor w/ 3 levels "1","2","3": 1 2 2 1 2 2 2 1 1 3 ...
 $ CTP              : Factor w/ 3 levels "1","2","3": 1 2 2 1 1 2 2 2 1 3 ...
 $ treat.method     : Factor w/ 3 levels "1","2","3": 2 1 2 1 1 1 2 3 1 1 ...
 $ MELD             : num  66.6 62 58.9 52.9 62.4 ...
 $ Gender           : int  1 1 1 1 0 0 1 0 0 1 ...
 $ Fundus.of.stomach: int  2 2 2 2 2 2 2 2 2 2 ...
 $ Glue             : int  0 0 0 0 0 0 0 0 1 0 ...
 $ age              : int  27 51 54 52 69 48 44 51 48 51 ...
 $ diameter         : int  0 0 0 0 0 0 0 0 0 0 ...
 $ WBC              : num  2.6 2.39 4.04 3.97 3.9 ...
 $ Hb               : int  94 138 99 158 129 117 88 53 117 114 ...
 $ PLT              : int  87 67 44 122 52 76 265 61 50 104 ...
 $ ALT              : int  29 26 10 337 13 29 13 20 8 41 ...
 $ AST              : int  29 23 16 167 39 47 28 28 11 57 ...
 $ GGT              : int  57 74 37 103 47 38 138 24 11 87 ...
 $ CHE              : int  4695 4815 3023 7997 3139 3612 3467 4163 4101 1828 ...
 $ INR              : num  1.42 1.22 1.19 1.07 1.36 1.07 1.27 1.35 1.74 1.42 ...
 $ CR               : num  92 89 82 37 66 47 45 37 43 70 ...
 $ Urea             : num  9.5 10.3 5 4.6 9 3.9 4.8 10.2 4.3 5.8 ...
 $ GLU              : num  5.65 5.88 7.14 6.92 7.67 5.53 5.84 7.36 9.82 5.49 ...
 $ Ammonia          : num  15.4 8.5 13.1 11.7 57.4 31.9 25.3 2.6 32.3 11.1 ...
 $ TBIL             : num  31 15.8 9.1 19.1 27.1 23.3 29.7 12.3 11.8 51.2 ...
 $ ALB              : num  47.5 36.1 35.4 52.7 38.5 33.4 37.6 46.6 40 29.1 ...

2 建立formula连接

> aa1 <- paste("recurrence",colnames(data[-1]),sep="~")
> aa1
 [1] "recurrence~Etiology"          "recurrence~Child.pugh"       
 [3] "recurrence~Ascites"           "recurrence~ALBI.stage"       
 [5] "recurrence~CTP"               "recurrence~treat.method"     
 [7] "recurrence~MELD"              "recurrence~Gender"           
 [9] "recurrence~Fundus.of.stomach" "recurrence~Glue"             
[11] "recurrence~age"               "recurrence~diameter"         
[13] "recurrence~WBC"               "recurrence~Hb"               
[15] "recurrence~PLT"               "recurrence~ALT"              
[17] "recurrence~AST"               "recurrence~GGT"              
[19] "recurrence~CHE"               "recurrence~INR"              
[21] "recurrence~CR"                "recurrence~Urea"             
[23] "recurrence~GLU"               "recurrence~Ammonia"          
[25] "recurrence~TBIL"              "recurrence~ALB"

3 建立空数据集

P_1 <- list()
combine_1 <- list()
combine_2 <- data.frame()
P_2 <- data.frame()

4 for循环

#for循环
for (i in 1:26) {
  fit_logistic1 <- glm(formula(aa1[i]),family = binomial(),data = data)
  
  P_1[[i]]<- data.frame(P=coef(summary(fit_logistic1))[,4])
  
  combine_1[[i]]<-data.frame(exp(cbind(OR=coef(fit_logistic1),confint(fit_logistic1))))

  #P_file <- paste0(paste("p",i,sep="_"),".csv")
  #write.table(P[[1]],file=paste0("20211101/1/",P_file,sep = ""))
  
  #OR_file <- paste0(paste("OR",i,sep="_"),".csv")
  #write.table(combine[[i]],file=paste0("20211101/1/",OR_file,sep = ""))
  
  combine_2 <- rbind(combine_2,combine_1[[i]])
  P_2 <- rbind(P_2,P_1[[i]])
}

5 合并及导出数据

#合并
combine_3 <- cbind(combine_2,P_2)

write.csv(combine_3,"20211101/1/combine_3.csv")