随机抽样以及数据探索

生信喵

一、随机抽样
       在做统计分析的过程中，经常需要进行随机抽样，R 提供了多种生成随机数的函数，并且可以进行多种形式的抽样。

1. x <- 1:10
   
2. x
   
3. sample(x,size = 5)
   
4. sample(x,size = 5,replace = F)
   
5. sample(x,size = 5,replace = T)
   
6. set.seed(1234)
   
7. sample(x,5,replace = F)#先设置上一步，这一步结果就固定了
   
8. set.seed(1234)
   
9. sample(x,4)
   
10. set.seed(1234)
    
11. sample(x,3)
    
12. #按比例抽样
    
13. x <- 1:100
    
14. sample(x,size = 0.25*length(x))

复制代码

二、利用 R 语言斗地主

1. rm(list = ls())
   
2. type <- c("red","spades","cube","plum")
   
3. amount <- c("A",2:10,"J","Q","K")
   
4. a <- rep(type,each = 13)
   
5. b <- rep(amount,4)
   
6. paste(a,b)
   
7. 
   
8. group <- expand.grid(type,amount)
   
9. poker <- paste(group$Var2,group$Var1,sep = "-")
   
10. poker
    
11. poker[c(53,54)] <- c("black Joker","red Joker")
    
12. set.seed(666)
    
13. shuffle <- sample(poker,54,replace = F)
    
14. head(shuffle) #运行种子数，每次牌一样
    
15. dipai <- shuffle[52:54]
    
16. shuffle <- shuffle[-c(52:54)]
    
17. one <- shuffle[c(T,F,F)]
    
18. two <- shuffle[c(F,T,F)]
    
19. three <- shuffle[c(F,F,T)]

复制代码

三、探索数据

1. rm(list = ls())
   
2. x <- read.csv('WHO.csv',row.names = 1)
   
3. x$CountryID <- rownames(x)
   
4. colnames(x)[grep('Pop',colnames(x))]
   
5. y <- data.frame(x$CountryID,x$Population_total)
   
6. sort(y$Population_total, decreasing = T)
   
7. order(y$Population_total, decreasing = T)
   
8. y[order(y$x.Population_total, decreasing = T),][1:10,]
   
9. sum(y$x.Population_total,na.rm = T)/100000000

复制代码