In the distribution below, how would I find the values that lies between 30% and 50% of the distribution? R code would be handy but not essential!
28, 28, 29, 30, 30, 31, 31, 31, 31, 32, 32, 32, 34, 35, 36, 37, 38
$\begingroup$
$\endgroup$
2
-
$\begingroup$ You need to specify better your question. Is it about probability or frequency? In the later case, a simple question of interpolation in the frequency empirical distribution. If you want inference, like confidence intervals, we need more. $\endgroup$– kjetil b halvorsen ♦Jun 22, 2015 at 8:16
-
$\begingroup$ Question edited. Don't need inference/confidence intervala $\endgroup$– lucianoJun 22, 2015 at 8:22
Add a comment
|
2 Answers
$\begingroup$
$\endgroup$
2
So you just want the quantiles of the sample! If you do not know what they are, look at wikipedia: https://en.wikipedia.org/?title=Quantile When specified by percent values, quantiles are called percentiles.
Here is some simple R code:
d <- c(28, 28, 29, 30, 30, 31, 31, 31, 31, 32, 32, 32, 34, 35, 36, 37, 38)
quantile(d,c(0.3, 0.5))
30% 50%
30.8 31.0
answered Jun 22, 2015 at 9:11
-
$\begingroup$ thanks, you've answered my question. But regarding terminology, I didn't want to quantiles, did I? The quantiles are 0.3 and 0.5, so I wanted the values at the point of these quantiles (whatever these values are called). $\endgroup$– lucianoJun 22, 2015 at 10:34
-
3
$\begingroup$
$\endgroup$
Since you specifically asked for the elements of the vector meeting your condition:
how would I find the values that lies between 30% and 50% of the distribution
test=c(28,28,29,30,30,31,31,31,31,32,32,32,34,35,36,37,38)
q.30=quantile(test,0.30)
q.50=quantile(test,0.50)
test[(test[]>=q.30)&((test[]<=q.50))]
