class notes April 13, 2010

# class notes April 13, 2010 - Treatment 1 B0 = 125 or 200...

This preview shows pages 1–3. Sign up to view the full content.

Treatment 1 B0 = 125 or 200 B1= 50,0 Treatment 2 B0= 125 or 200 B1= -50,0 Regression results in file Project540 Unconditional mean of the sample containing y1,y2,…. Is ybar= E(y) What if we have info about other vars that I think affect y? Conditional mean Ybar c= E(Y given X) If x1 and x2 are useful in explaining variation in y, then either a. b1 ne 0, b2 ne 0 b. b1 = 0, B2 ne 0 c. b1 ne 0, b2 e 0 If both B1= 0, and B2 = 0, E(Y given X) = B0 = E(y) If either or both have info about y, then should be able to reject H0: B1 = B2 =0 versus alternative that Ha: at least one ne to 0. F = (R^2/k)/(1-R2)/(n-k-1) Alternatively, could test with two separate t-tests: H0: B0 = 0, H0: B1= 0 a. But f statistic confidence interval is smaller. It’s harder to reject H0 because of this. Y bar = sample mean Yj hat= predicted value from the regression Prediction error= Yj – Ybar in the case of the mean

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

View Full Document
Predicition error = Yj – yhat for the regression
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 06/10/2010 for the course HIS 102 taught by Professor Archer during the Spring '10 term at University of Mississippi Medical Center.

### Page1 / 3

class notes April 13, 2010 - Treatment 1 B0 = 125 or 200...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online