-3

I have a vector:

vector <- c(12, 17, 24, 35, 23, 34, 56)

How to calculate confidence intervals (90%, 99%, 95%) for this vector in R?

This is example of result I want: enter image description here

Cuong.S
  • 39
  • 1
  • 1
  • 5
  • Confidence intervals can be defined/calculated for a specific metric (population parameter) you have in mind (such as the mean) and not for a set of values (such as a vector). – AntoniosK Feb 04 '18 at 20:02

1 Answers1

8

Here is a function that will calculate your confidence interval according to the t-distribution:

confidence_interval <- function(vector, interval) {
  # Standard deviation of sample
  vec_sd <- sd(vector)
  # Sample size
  n <- length(vector)
  # Mean of sample
  vec_mean <- mean(vector)
  # Error according to t distribution
  error <- qt((interval + 1)/2, df = n - 1) * vec_sd / sqrt(n)
  # Confidence interval as a vector
  result <- c("lower" = vec_mean - error, "upper" = vec_mean + error)
  return(result)
}

And example usage for your provided vector and intervals:

> vector <- c(12, 17, 24, 35, 23, 34, 56)
> confidence_interval(vector, 0.90)
   lower    upper 
17.97255 39.45602 
> confidence_interval(vector, 0.95)
   lower    upper 
15.18797 42.24060 
> confidence_interval(vector, 0.99)
    lower     upper 
 8.219946 49.208626

And this is the tutorial from which I developed this method.

Eugene Brown
  • 4,032
  • 6
  • 33
  • 47