Question

I have the example data frame

test.df<-data.frame(classifier=runif(n=1000), x1=rnorm(1000), x2=rnorm(1000), x3=rnorm(1000))

with x1,x2,…,x10000

I would like to use the apply function to perform a large amount of tests (lets say t.test) and collect the results in a vector(lets say t.test()$p.value ). The test on a single column would be t.test(test.df$x1[ test.df$classifier<0.4 ], test.df$x1[ test.df$classifier>0.6 ])$p.value

I would like to perform this on all with x1,x2,…,x10000. I though I would use the apply function with MARGIN=2 but I cannot get the apply function to split every column vector according to the classifier.

(The example provided above has absolutely no statistical meaning. Don’t go viral)

Any help?

Answer 1

Is this what you want ?

df<-data.frame(classifier=runif(n=1000), x1=rnorm(1000), x2=rnorm(1000), x3=rnorm(1000))
sapply(df[,-1], function(v) {
  t.test(v[df$classifier<0.4], v[df$classifier>0.6])$p.value
})

Which gives :

       x1        x2        x3 
0.5028683 0.1238735 0.2021623