分组统计
sd=aggregate(df$Var,by=list(df$group),FUN=sd)
mean=aggregate(df$Var,by=list(df$group),FUN=mean)
ANOVA
anova=aov(df$Var1~df$Var2)
summary(anova)
- 这的的df就是原始数据
- Var1是需要比较的变量,Var2是分组变量(factors)
Chi-square test
LowTIR=c(6,4,5,5)
HighTIR=c(26,17,53,29)
data_drug=data.frame(LowTIR,HighTIR,row.names = c("胰岛素","磺脲类","二甲双胍","阿卡波糖"))
![在这里插入图片描述](https://i-blog.csdnimg.cn/blog_migrate/38e75365012f067592184fa25d2478cb.png#pic_center)
Kruskal-Wallis Rank Sum Test
kruskal.test(nonnormVar~group,data = df)
正态性检验
qq图
qqnorm(x)
qqline(x)
Kolmogorov-Smirnov检验
ks.test(x,"pnorm")
library(nortest)
lillie.test(x)
Shapiro-Wilk检验
shapiro.test(x)
控制混杂
library(MatchIt)
成组匹配
个体匹配
PSM