Ch13 Multiple Regression

The model y = b 0 + b 1x1 + b 2x2 + e is a second-order regression model.

• True

• False

In the multiple regression model y = b 0 + b 1x1 + b 2x2 + b 3x3 + e, the b coefficients of the x variables are called partial regression coefficients.

• True

• False

In the model y = b 0 + b 1x1 + b 2x2 + b 3x3 + e, e is a constant.

• True

• False

The F test is used to determine whether the overall regression model is significant.

• True

• False

In a multiple regression analysis with N observations and k independent variables, the degrees of freedom for the residual error is given by (N – k).

• True

• False

A cost accountant is developing a regression model to predict the total cost of producing a batch of printed circuit boards as a linear function of batch size (the number of boards produced in one lot or batch), production plant (Kingsland and Yorktown), and production shift (day and evening). The response variable in this model is ___.

• batch size

• production shift

• production plant

• total cost

A market analyst is developing a regression model to predict monthly household expenditures on groceries as a function of family size, household income, and household neighbourhood (urban, suburban, and rural). The response variable in this model is ___.

• family size

• expenditures on groceries

• household income

• household neighbourhood

The number of degrees of freedom for error is ___.

• 1

• 4

• 34

• 30

49. In regression analysis, outliers may be identified by examining the ___.

a) coefficient of determination

b) coefficient of correlation

c) p values for the partial coefficients

d) residuals

e) R-squared value

Answer: d

• coefficient of determination

• coefficient of correlation

• residuals

• R-squared value

These results indicate that ___.

• none of the predictor variables are significant at the 5% level

• each predictor variable is significant at the 5% level

• x₁ is the only predictor variable significant at the 5% level

• x₂ is the only predictor variable significant at the 5% level