2/19/2009

Tying up Loose Ends

• Homework #2 & #3
• The normal correlation model and the bivariate normal (#2.42 & 2.45)
• When would an additional variable be useful? (#3.3f)
• Use of standardized residuals
• How to omit values in SAS?
• SSLF = SSE – SSPE
• Remember to:
• Interpret intervals
• Make conclusions in context

Chapter 6: Multiple Regression I
“Oh, this is why I needed to know Linear Algebra.”

• First-order model with p-1 predictor variables
• Interpretation of regression coefficients
• General linear regression model…
• …with qualitative predictor variables
• …with polynomial regression
• …with interaction effects

• Estimation of Regression coefficients
• Least squares criterion
• Typo on p. 223!
• ANOVA -- Differences from simple model – See Table 6.1
• Degrees of freedom
• F test
• Coefficient of multiple determination
• Inferences for regression parameters

• Diagnostics
• Scatterplot matrix
• Correlation matrix
• (Standardized) Residual Plots
• Tests
• F test for Lack of Fit

• Example - Exercise 6.30 on p. 254 (see senic.sas)

Reminder: Exam #1 is Tuesday!

There will be a problem session next Thursday over Chapter 6.