CODEWITHSUNDEEP
rdataframecalculated-columns
Alan
Alan

Reputation: 681

R Newbie...calculating averages

I'm very new to R and I'm still quite close to the bottom of the very steep learning curve I think. So...

I have a data frame (imported from a .csv file). Contains a number of fields - let's call them Field1, Field2, Field3... Field 10.

The fields are numeric. For each row of data I'd like to calculate the average of the lowest 3 (say) numbers. In other words:

((smallest number) + (second smallest number) + (third smallest number))/3

Also there are some NAs in the data so I'd like the calculation to return NA if there aren't actually three numbers to avarage over (although perhaps R will do this naturally anyway).

Is there a succinct way to do this in R and (better still) to store the result as a new field in the existing data frame?

Grateful for any advice. Thanks. A

Upvotes: 0

Views: 71

Answers (2)

paoloeusebi
paoloeusebi

Reputation: 1086

 data(mtcars) 

 avg3 <- sapply(mtcars, function(x) mean(head(sort(x[!is.na(x)]), 3))) 

 avg3

 mtcars.avg3 <- rbind(mtcars, avg3) 

 mtcars.avg3

Upvotes: 0

alexjackson
alexjackson

Reputation: 22

For each row of data I'd like to calculate the average of the lowest 3 (say) numbers.

I think this should achieve what you're asking:

data <- mtcars
data$low.mean <- apply(data, 1, function(x) mean(head(sort(x[!is.na(x)]), n = 3)))

Upvotes: 0

Related Questions