CODEWITHSUNDEEP
linuxrawk
moadeep
moadeep

Reputation: 4098

Print standard deviation of column in file using R/awk

I have a file with a column of numbers

4.685
5.440
5.751
4.685
3.979

In my shell script I would like to interrogate many of these files and get the standard deviation and mean

I can achieve mean using awk easily enough awk '{sum+=$1} END { print sum/NR}' file

When using awk for standard deviation awk '{x[NR]=$0; s+=$0} END{a=s/NR; for (i in x){ss += (x[i]-a)^2} sd = sqrt(ss/NR); print sd}' file

I get 0.625. This number differs from excel which gives me 0.699. I have since discovered I can execute R from the command line to print out the sd: R -q -e "x <- read.csv('file', header = F); sd(x[ , 1])"

However, this gives a slightly messy output

[1] 4.908
\> 
\>

Can I adjust the R command to print out only the number without resorting to head and cut/awk?

Also what is wrong with my awk code for extracting standard deviation?

Upvotes: 0

Views: 448

Answers (1)

keegan
keegan

Reputation: 2992

I can't quite tell what's wrong with your awk, but for the R command, you might find that write helps:

R -q -e "x<- read.csv('file.csv',header=FALSE)[,1] ; write(sd(x),file='result.txt')"

Upvotes: 0

Related Questions