Answer the following questions based on Section 4.3.2: Qualitative predictors with three or more levels.

Consider the following data:

X 0.8 0.6 0.8 0.2 0.5 1.0 0.3 0.1 0.2 0.5
D1 0 0 1 1 1 0 0 0 0 0
D2 0 0 0 0 0 1 0 0 1 1
Y 1.0 0.8 2.2 1.4 1.7 3.4 0.3 0.4 1.5 1.9
D1 and D2 are indicator variables representing a qualitative predictor with three levels (e.g., D1 = D2 = 0 represents “low,” D1 = 1 and D2 = 0 represents “medium,” and D1 = 0 and D2 = 1 represents “high”). Use statistical software to fit a multiple linear regression model with response variable, Y, and predictor variables, D1, D2, X, D1X, and D2X (the last two are interaction variables). Use the fitted model to answer the following questions.

1. 

Which predictor variables have p-values for their regression parameters that are less than a significance level of 0.05 (select all that apply)?

2. 

Fit a second multiple linear regression model omitting the predictor variable with the largest p-value. For this second model, which predictor variables have p-values for their regression parameters that are less than a significance level of 0.05 (select all that apply)?

3. 

This fitted model represents three regression lines for low (D1 = D2 = 0), medium (D1 = 1 and D2 = 0), and high (D1 = 0 and D2 = 1). Which of the following best describes these three fitted lines?

4. 

Which expression represents the predicted (or fitted) values of Y with respect to X for low (D1 = D2 = 0)?

5. 

Which expression represents the predicted (or fitted) values of Y with respect to X for medium (D1 = 1 and D2 = 0)?

6. 

Which expression represents the predicted (or fitted) values of Y with respect to X for high (D1 = 0 and D2 = 1)?

7. 

Compare s, R2, and adjusted R2 for the second model to the first model.

Categories:

Tags:

Comments are closed