# confsum - Statistics Confidence intervals summary sheet...

This preview shows pages 1–2. Sign up to view the full content.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Statistics: Confidence intervals summary sheet Introduction • Suppose that we take a SRS of size n from a population with mean μ and variance σ 2 . We know that the probability distribution of the sample mean ¯ X is at least approximately ¯ X ∼ N ( μ,σ 2 /n ) for sufficiently large n . Thus we can make statements of probability about where ¯ X is likely to be. How does this help us decide where μ is likely to be? • The standardised version of the sample mean is ¯ X- μ σ/ √ n ∼ N (0 , 1). Thus we can determine from tables that P ( ¯ X- μ σ √ n ≤ 1 . 96) = 0 . 975 , P (- 1 . 96 < ¯ X- μ σ √ n ≤ 1 . 96) = 0 . 95 . • This probability statement can be rearranged into P ( ¯ X- 1 . 96 σ √ n < μ ≤ ¯ X + 1 . 96 σ √ n ) = 0 . 95 , so that μ lies in the interval ( ¯ X- 1 . 96 σ √ n , ¯ X +1 . 96 σ √ n ) with probability 95%. This probability statement is about μ rather than ¯ X . However it is useless as a tool for inference because ¯ X here is a random variable – remember: probability calculations concern random variables, and we cannot replace ¯ X by one of its observed values here and still call it a probability statement. • Suppose that we do replace ¯ X by ¯ x , its observed value calculated from data. What is the status of the interval (¯ x- 1 . 96 σ √ n , ¯ x + 1 . 96 σ √ n )? It is called a confidence interval for μ . This confidence interval is at 95% because the theoretical probability statement from which the confidence interval evolved had probability 0.95....
View Full Document

### Page1 / 4

confsum - Statistics Confidence intervals summary sheet...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online