conditional elseif statements in R

Question

I have a data framethat looks like this Animal species sum A 2 A 6 B 8 B 1 C 6 C 3 D 5 D 4 I wants the code that divides the min sum by the max sum for each species and creates a new column called ratio, and if that number is greater than .2 I want it to return that ratio, but if it is smaller than .2, I want it to return NA. I would like to do this in dplyr This is the code I have currently but it is not working I have this string of commands in R animal <- animal %>% + group_by(species) %>% + mutate(ratio = ifelse((min(sum)/max(sum) > .2), (min(sum)/max(sum)), "NA")) Thanks! This is what it should look like in the end species sum ratio A 2 .333 A 6 .333 B 8 NA B 1 NA C 6 .5 C 3 .5 D 5 .8 D 4 .8

Answer 1

I would avoid ifelse because both you will have to evaluate the whole vector anyway, and because you need to calculate the whole process twice. Here's how I would do this library(data.table) setDT(Animal)[, ratio := {r <- range(sum); r[1L]/r[2L]}, by = species] Animal[ratio <= .2, ratio := NA] # species sum ratio # 1: A 2 0.3333333 # 2: A 6 0.3333333 # 3: B 8 NA # 4: B 1 NA # 5: C 6 0.5000000 # 6: C 3 0.5000000 # 7: D 5 0.8000000 # 8: D 4 0.8000000

Answer 2

You can use NA_real_ to make the types compatible animal %>% group_by(species) %>% mutate(ratio= ifelse((min(sum)/max(sum))> .2, round((min(sum)/max(sum)),2), NA_real_)) # species sum ratio #1 A 2 0.33 #2 A 6 0.33 #3 B 8 NA #4 B 1 NA #5 C 6 0.50 #6 C 3 0.50 #7 D 5 0.80 #8 D 4 0.80 A base R option would be animal$ratio <- with(animal, ave(sum, species, FUN=function(x) { x1 <- min(x)/max(x) NA^(x1 <= 0.2)*x1 }))

Answer 3

animal <- animal %>% group_by(species) %>% mutate(ratio = ifelse((min(sum)/max(sum) > .2), (min(sum)/max(sum)), as.numeric(NA))) > animal Source: local data frame [8 x 3] Groups: species species sum ratio 1 A 2 0.3333333 2 A 6 0.3333333 3 B 8 NA 4 B 1 NA 5 C 6 0.5000000 6 C 3 0.5000000 7 D 5 0.8000000 8 D 4 0.8000000