Question 13

What is the main purpose of carrying out a residual or error analysis in a regression analysis?

Question 13 options:

	To estimate the regression coefficients.
	To determine whether the errors are normally distributed.
	To test for the validity of the regression model.
	To measure the prediction error.
	To determine whether the mean of the residuals is 0.

Question 14

. A multiple linear regression analysis was carried out using the following data for 48 months of sales of ski equipment for a US company.

Y = Sales (in millions of dollars).

X₁ = Advertising expenditure (in millions of dollars )

X₂ = 1 if the month is in the high season

= 0 if the month is not in the high season

X₃ = X₁ X₂

The results of a regression analysis used to predict sales are as follows:

1. Prediction equation: Y = 50 + 1.20 X₁ + 20X₂ + 0.62X₃

2. The F test is statistically significant.

3. The three T tests are statistically significant.

4. The value of SSR = 16,016.

5. The value of SST = 17,600.

6. The residual plots exhibited a random pattern with no outliers present

The 50^th month is a low season month with an advertising expenditure of $5 million. The 56^th month is a high season month with an advertising expenditure of $10 million. Estimate the difference in sales between the 56^th month and the 50^th month to the nearest million dollars.

Question 14 options:

	$9 million
	$32 million.
	$6 million
	$3 million
	$29 million.