Class_18

Class_18 - The Pearson Product-Moment Correlation...

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon
The Pearson Product-Moment Correlation Coefficient
Background image of page 1

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

View Full DocumentRight Arrow Icon
The regression coefficient is an asymmetrical statistic, one that gives different values for the model Y = f(X) and the model X = f(Y). The other major measure of bivariate association is the Pearson product-moment correlation coefficient (sometimes called "little r" for short). The correlation coefficient is a symmetrical statistic. That is, it simply describes the association between X and Y without worrying about whether Y = f(X) or X = f(Y). It would produce the same result in either case. Unlike the regression coefficient, whose values range from 0.0 to ± , the correlation coefficient ranges from 0.0 when there is NO association between X and Y to ± 1.00 when there is PERFECT association (either direct or inverse).
Background image of page 2
To generate the second set of statistics describing association from the linear model, we partition the sum of squares . Graphically, we begin with a single data point, i, in two-dimensional space. Y i is its location on the scale of y (on the y-axis); below that is the predicted location of Y, Y i -hat. The dotted horizontal line (- - - -) is the location of the mean of Y. (When there is no association between X and Y, b = 0.0 and therefore a = Y-bar.) where b = 0, X b Y a + = Y a =
Background image of page 3

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

View Full DocumentRight Arrow Icon
i Y i } Y i - hat _ } Y - - - - - - - - - - - - - - - X i ( 29 i i Y Y ˆ - ( 29 Y Y i - ˆ
Background image of page 4
Background image of page 5

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

View Full DocumentRight Arrow Icon
The vertical line represents the deviation of the i th observation from the mean of Y (i.e., the difference between Y i and Y-bar). The line of best fit bisects the deviation into its two mathematical components. The component ABOVE the line of best fit is the residual , the difference between Y i and Y i - hat, the actual location of the i th observation on the y-axis and the predicted location of this observation on the y-axis. This is the error (or residual) component .
Background image of page 6
The component BELOW the line of best fit is new. It is
Background image of page 7

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

View Full DocumentRight Arrow Icon
Image of page 8
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 02/04/2008 for the course PPD 404 taught by Professor Velez during the Fall '07 term at USC.

Page1 / 22

Class_18 - The Pearson Product-Moment Correlation...

This preview shows document pages 1 - 8. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online