310projectlastyear

310projectlastyear - 109) Since circulation has p-value...

Info iconThis preview shows pages 1–5. Sign up to view the full content.

View Full Document Right Arrow Icon
Background image of page 1

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

View Full DocumentRight Arrow Icon
Background image of page 2
1)a) Pagecost and circulation seem to be pretty right skewed. Percentage of male and median income seem to be bimodal and do not really have any pattern of skewness. b) y=bo + b1x1 + b2x2 + b3x3 + Epsilon
Background image of page 3

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

View Full DocumentRight Arrow Icon
on
Background image of page 4
B) Page cost and circulation seem to have a moderately strong curve relationship close to natural log relationship. There seems to be no clear relationship between Page cost and the other two explanatory variables. 2) Residual diagnostic because Epsilon is estimated by residuals. a) Since p value for the F-test is 0.000, so YES, the model is useful in predicting the response. b) y= 7008 + 3.96x1 – 34.7x2 + .7x3 y= pagecost x1= circulation x2= percentage of males x3= median income c) Circulation is very strongly significant (t-test , p-value is .000), percentage of males is not significant (t-test, p value is .815). Median income is also not significant ( t-test, p value is .
Background image of page 5
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 109) Since circulation has p-value less than .05, I would keep it in the model. Percentage of males, however, should first be taken out since it is the most insignificant variable due to its highest p-value of .815. If necessary, I would remove median income as well. d)scatterplot of residuals versus predicted values there is a certain pattern present in the residual plot, hence the linearity assumption is not satisfied. Perhaps the relationship is not linear. Heteroscedasticity does not seem as evident here. A transformation applied to the variables may fix these problems. For example, I would probably try to transform the page cost and or circulation variables by taking the natural log. e)...
View Full Document

This note was uploaded on 09/11/2011 for the course BUAD 310 taught by Professor Lv during the Spring '07 term at USC.

Page1 / 5

310projectlastyear - 109) Since circulation has p-value...

This preview shows document pages 1 - 5. Sign up to view the full document.

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