- Get subset from Vector
- [] return an object of the same classes as extracted object
> x <- 1:10
> x[x>5] # extract subset which includes
[1] 6 7 8 9 10
> x[2]
[1] 2
> x[3]
[1] 3
Get 1,3,5th elements from x
> x[c(1,3,5)]
[1] 1 3 5
What if we want to get all elements except 1,3,5th elements?
> x[c(-1,-3,-5)] # or simply use x[-c(1,3,5)]
[1] 2 4 6 7 8 9 10
Get subset from Frame
> x
var1 var2 var3
1 3 10 11
2 2 6 15
3 4 8 12
4 5 7 14
5 1 9 13
> x[,1] # get first column data
[1] 3 2 4 5 1
> x[,"var3"] # get var3 column data
[1] 11 15 12 14 13
> x[2:3,"var3"] # get 2,3rd data in column var3
[1] 15 12
> x[(x$var2 >8 & x$var3 <15),] # get all column data, with 2nd data in column var2 >8 and 3rd in column var3 <15
var1 var2 var3
1 3 10 11
5 1 9 13
Example for dealing with NA values
> x$var1[c(1,4)] = NA
> x
var1 var2 var3
3 NA 8 12
5 1 9 13
2 2 6 15
4 NA 7 14
1 3 10 11
> x[x$var1>1,] # lines with NA will show up as well
var1 var2 var3
NA NA NA NA
2 2 6 15
NA.1 NA NA NA
1 3 10 11
> x[which(x$var1>1),] # use which to ignore NA line
var1 var2 var3
2 2 6 15
1 3 10 11
Example of sorting
> sort(x$var1)
[1] 1 2 3
> sort(x$var1, decreasing=TRUE)
[1] 3 2 1
> sort(x$var1, na.last = TRUE)
[1] 1 2 3 NA NA
Example of reordering frame
> x[order(x$var2),]
var1 var2 var3
2 2 6 15
4 NA 7 14
3 NA 8 12
5 1 9 13
1 3 10 11
> library(plyr) # use plyr package
> arrange(x,var3)
var1 var2 var3
1 3 10 11
2 NA 8 12
3 1 9 13
4 NA 7 14
5 2 6 15
> arrange(x,desc(var3)
+ )
var1 var2 var3
1 2 6 15
2 NA 7 14
3 1 9 13
4 NA 8 12
5 3 10 11
Example of adding column
> x$var4 <- rnorm(5)
> x
var1 var2 var3 var4
3 NA 8 12 0.01046482
5 1 9 13 0.06659688
2 2 6 15 0.91059308
4 NA 7 14 1.26587778
1 3 10 11 -1.46815620
> y <- cbind(x, rnorm(5))
> y
var1 var2 var3 var4 rnorm(5)
3 NA 8 12 0.01046482 0.4359661
5 1 9 13 0.06659688 -0.7918177
2 2 6 15 0.91059308 -0.0485241
4 NA 7 14 1.26587778 -0.5299538
1 3 10 11 -1.46815620 0.1181559
Get subset from Matrix
- Example for getting one element from matrix and one row from it
> x <- matrix(1:6,2,3)
> x
[,1] [,2] [,3]
[1,] 1 3 5
[2,] 2 4 6
> x[1,2] # get element in first row, second column
[1] 3
> x[1,] # get first row
[1] 1 3 5
It will get vector by default, you could also get matrix by using drop argument
> x[1,2,drop = FALSE]
[,1]
[1,] 3
Get subset from List
- Example for getting subset from List
> x <- list(male=c(1:4),female=c(5:10)) # create a list
> x # print
$male
[1] 1 2 3 4
$female
[1] 5 6 7 8 9 10
> x[1] # single bracket return list
$male
[1] 1 2 3 4
> x[[1]] # double bracket return vector
[1] 1 2 3 4
> x$male # use name to return vector
[1] 1 2 3 4
Use variable indice to get subset
> var <- "male"
> x[[var]]
[1] 1 2 3 4
Deal with NA values
- Get complete cases over two vector
> x <- c("male","female",NA,"male")
> y <- c("female",NA,NA,"female")
> completeflag <- complete.cases(x,y) # save complete flag TRUE,FALSE,FALSE,TRUE
> x[completeflag]
[1] "male" "male"
Remove NA from vector
> x <- c(1:10,NA)
> x
[1] 1 2 3 4 5 6 7 8 9 10 NA
> naflag <- is.na(x)
> x[naflag]
[1] NA
> x[!naflag]
[1] 1 2 3 4 5 6 7 8 9 10
Sample data from vector(s)
- Use sample() function to get number of samples from target
> x <- c(1:10)
> y <- c("a","b","c","d")
> sample(c(x,y),4)
[1] "3" "10" "2" "c"
How to count the number of NA values?
- R treats TRUE as 1 and FALSE as 0, so we could check sum of the vector to see the count.
> x <- rep(NA, 10)
> y <- c(1:10)
> z <- sample(c(x,y),5)
> z
[1] 9 NA 8 NA NA
> naflag <- is.na(z)
> naflag
[1] FALSE TRUE FALSE TRUE TRUE
> sum(naflag)
[1] 3
No comments:
Post a Comment