CODEWITHSUNDEEP
rif-statementdplyrgroup-byconditional-statements
Noel
Noel

Reputation: 109

Creating a new column based on multiple conditional statements in r

I have n numbers ranging from 0 to 0.6; I want to create a new column grouping this values with multiple conditional statements. If a number is between 0 - 0.1; name it = 0.1; from 0.1 - 0.2, group= 0.2; from 0.2 - 0.3, group=0.3; and so on.

value<- runif(20, min=0, max=0.6)
df<- data.frame(value)

I tried this way, but is not working, it is just assigning a correct answer for the first value, and not to the rest.

#Assigning groups for values
for (i in seq(1, length(df$value))) {
  if (df$value[i] > 0 & df$value[i] <= 0.1 ){
    df$group[i]<- "0.1"
  } else if (df$value[i] > 0.1 & df$value[i] <= 0.2 ){
    df$group[i]<- "0.2"
  } else if (df$value[i] > 0.2 & df$value[i] <= 0.3 ){
    df$group[i]<- "0.3"
  } else if (df$value[i] > 0.3 & df$value[i] <= 0.4 ){
    df$group[i]<- "0.4"
  } else if (df$value[i] > 0.4 ){
    df$group[i]<- "0.5" 
  }
return(df) 
}

Any idea or a better way to do this. Thanks in advance.

Upvotes: 0

Views: 102

Answers (2)

Dan Kennedy
Dan Kennedy

Reputation: 430

The cut function can convert the numeric vector into a factor:

df$group = cut(df$value,
           breaks = c(0,0.1,0.2,0.3,0.4,0.5,0.6),
           labels = c('0.1','0.2','0.3','0.4','0.5','0.6'))
head(df)
#      value group
#1 0.4204752   0.5
#2 0.4691363   0.5
#3 0.3926192   0.4
#4 0.0418944   0.1
#5 0.1074853   0.2
#6 0.1914169   0.2
levels(df$group)
#[1] "0.1" "0.2" "0.3" "0.4" "0.5" "0.6"

Upvotes: 1

Ronak Shah
Ronak Shah

Reputation: 388817

You can use plyrs round_any function

df$value1 <- plyr::round_any(df$value, 0.1, ceiling)
df
#        value value1
#1  0.59465953    0.6
#2  0.10581043    0.2
#3  0.48806113    0.5
#4  0.04106798    0.1
#5  0.24026985    0.3
#6  0.08468660    0.1
#7  0.11598592    0.2
#8  0.50481103    0.6
#9  0.43194839    0.5
#10 0.16032725    0.2
#11 0.29700099    0.3
#12 0.04986834    0.1
#13 0.21233054    0.3
#14 0.58152528    0.6
#...

Upvotes: 3

Related Questions