# ch8 - STAT1306 Introductory Statistics Chapter 8...

This preview shows pages 1–7. Sign up to view the full content.

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

View Full Document

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

View Full Document

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: STAT1306 Introductory Statistics Chapter 8 Correlation and Regression Department of Statistics and Actuarial Science The University of Hong Kong Prof. W.K. Li (hrntlwk@hku.hk) STAT1306 (Chapter 8) 2008-09 1st Semester 1 / 25 Correlation Recall in Chapter 5 we introduced the concept of covariance and correlation between two random variables X and Y : covariance : Cov ( X, Y ) = E ( XY )- E ( X ) E ( Y ) correlation coefficient: ρ ( X, Y ) = Cov ( X, Y ) σ X σ Y = ρ for simplicity where σ X and σ Y are the standard deviations of X and Y respectively. Note that ρ does not depend on the scales of X and Y and- 1 ≤ ρ ≤ 1. Prof. W.K. Li (hrntlwk@hku.hk) STAT1306 (Chapter 8) 2008-09 1st Semester 2 / 25 The correlation coefficient ρ measures the strength of a linear relationship with ρ = 0 implies no such relationship and ρ = ± 1 a perfect linear relationship. Given observations of ( X, Y ): ( X 1 , Y 1 ) , ··· , ( X n , Y n ) we can estimate ρ by b ρ = S XY √ S XX √ S Y Y where S XY = 1 n n X i =1 ( X i- X )( Y i- Y ) , S XX = 1 n n X i =1 ( X i- X ) 2 with a similar definition for S Y Y . Prof. W.K. Li (hrntlwk@hku.hk) STAT1306 (Chapter 8) 2008-09 1st Semester 3 / 25 Hence, b ρ = n ∑ i =1 ( X i- X )( Y i- Y ) r n ∑ i =1 ( X i- X ) 2 · n ∑ i =1 ( Y i- Y ) 2 (some text books use r to denote b ρ ) Prof. W.K. Li (hrntlwk@hku.hk) STAT1306 (Chapter 8) 2008-09 1st Semester 4 / 25 Prof. W.K. Li (hrntlwk@hku.hk) STAT1306 (Chapter 8) 2008-09 1st Semester 5 / 25 It is often of interest to know if ρ = 0? That is, we want to test H : ρ = 0 vs H 1 : ρ 6 = 0 ....
View Full Document

## This note was uploaded on 09/06/2010 for the course STAT STAT1306 taught by Professor Prof during the Fall '08 term at HKU.

### Page1 / 25

ch8 - STAT1306 Introductory Statistics Chapter 8...

This preview shows document pages 1 - 7. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online