• Home
  • Textbooks
  • Statistics for Business and Economics: Global Edition
  • Multiple Variable Regression Analysis

Statistics for Business and Economics: Global Edition

Newbold P., Carlson W.L., Thorne B.M.

Chapter 12

Multiple Variable Regression Analysis - all with Video Answers

Educators


Chapter Questions

03:04

Problem 1

Given the estimated linear model
$$
\hat{y}=10+3 x_1+2 x_2+4 x_3
$$
a. Compute $\hat{y}$ when $x_1=20, x_2=11$, and $x_3=10$.
b. Compute $\hat{y}$ when $x_1=15, x_2=14$, and $x_3=20$.
c. Compute $\hat{y}$ when $x_1=35, x_2=19$, and $x_3=25$.
d. Compute $\hat{y}$ when $x_1=10, x_2=17$, and $x_3=30$.

Diwakar Mandilwar
Diwakar Mandilwar
Numerade Educator
03:04

Problem 2

Given the estimated linear model
$$
\hat{y}=10+5 x_1+4 x_2+2 x_3
$$
a. Compute $\hat{y}$ when $x_1=20, x_2=11$, and $x_3=10$.
b. Compute $\hat{y}$ when $x_1=15, x_2=14$, and $x_3=20$.
c. Compute $\hat{y}$ when $x_1=35, x_2=19$, and $x_3=25$.
d. Compute $\hat{y}$ when $x_1=10, x_2=17$, and $x_3=30$.

Diwakar Mandilwar
Diwakar Mandilwar
Numerade Educator
03:04

Problem 3

Given the estimated linear model
$$
\hat{y}=10+2 x_1+12 x_2+8 x_3
$$
a. Compute $\hat{y}$ when $x_1=20, x_2=11, x_3=10$.
b. Compute $\hat{y}$ when $x_1=15, x_2=24, x_3=20$.
c. Compute $\hat{y}$ when $x_1=20, x_2=19, x_3=25$.
d. Compute $\hat{y}$ when $x_1=10, x_2=9, x_3=30$.

Diwakar Mandilwar
Diwakar Mandilwar
Numerade Educator
03:51

Problem 4

Given the following estimated linear model
$$
\hat{y}=10+2 x_1+12 x_2+8 x_3
$$
a. What is the change in $\hat{y}$ when $x_1$ increases by 4 ?
b. What is the change in $\hat{y}$ when $x_3$ increases by 1 ?
c. What is the change in $\hat{y}$ when $x_2$ increases by 2 ?

Diwakar Mandilwar
Diwakar Mandilwar
Numerade Educator

Problem 4

An aircraft company wanted to predict the number of worker-hours necessary to finish the design of a new plane. Relevant explanatory variables were thought to be the plane's top speed, its weight, and the number of parts it had in common with other models built by the company. A sample of 27 of the company's planes was taken, and the following model was estimated:
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\beta_3 x_3+\varepsilon
$$
where
$y=$ design effort, in millions of worker-hours
$x_1=$ plane's top speed, in miles per hour
$x_2=$ plane's weight, in tons
$x_3=$ percentage number of parts in common with other models

The estimated regression coefficients were as follows:
$$
b_1=0.661 \quad b_2=0.065 \quad b_3=-0.018
$$
and the estimated intercept was 2.0.
Predict design effort for a plane with a top speed of Mach 1.0, weighing 7 tons, and having $50 \%$ of its parts in common with other models.

Check back soon!
04:17

Problem 5

Given the following estimated linear model
$$
\hat{y}=10-2 x_1-14 x_2+6 x_3
$$
a. What is the change in $\hat{y}$ when $x_1$ increases by 4 ?
b. What is the change in $\hat{y}$ when $x_3$ decreases by 1 ?
c. What is the change in $\hat{y}$ when $x_2$ decreases by 2 ?

Diwakar Mandilwar
Diwakar Mandilwar
Numerade Educator

Problem 6

An aircraft company wanted to predict the number of worker-hours necessary to finish the design of a new plane. Relevant explanatory variables were thought to be the plane's top speed, its weight, and the number of parts it had in common with other models built by the company. A sample of 27 of the company's planes was taken, and the following model was estimated:
$$
y_i=\beta_0+\beta_1 x_{1 i}+\beta_2 x_{2 i}+\beta_3 x_{3 i}+\varepsilon_i
$$
where
$y_i=$ design effort, in millions of worker-hours
$x_{1 i}=$ plane's top speed, in miles per hour
$x_{2 i}=$ plane's weight, in tons
$x_{3 i}=$ percentage number of parts in common with other models

The estimated regression coefficients were as follows:
$$
b_0=2 \quad b_1=0.661 \quad b_2=0.065 \quad b_3=-0.018
$$
Interpret these estimates.

Check back soon!
01:41

Problem 7

In a study of the influence of financial institutions on bond interest rates in Germany, quarterly data over a period of 12 years were analyzed. The postulated model was
$$
y_i=\beta_0+\beta_1 x_{1 i}+\beta_2 x_{2 i}+\varepsilon_i
$$
where
$y_i=$ change over the quarter in the bond interest rates
$x_{1 i}=$ change over the quarter in bond purchases by financial institutions
$x_{2 i}=$ change over the quarter in bond sales by financial institutions
The estimated regression coefficients were as follows:
$$
b_1=0.057 \quad b_2=-0.065
$$
Interpret these estimates.

Sheryl Ezze
Sheryl Ezze
Numerade Educator

Problem 8

The following model was fitted to a sample of $30 \mathrm{fami}-$ lies in order to explain household milk consumption:
$$
y_i=\beta_0+\beta_1 x_{1 i}+\beta_2 x_{2 i}+\varepsilon_i
$$
where
$$
\begin{aligned}
y_i & =\text { milk consumption, in quarts per week } \\
x_{1 i} & =\text { weekly income, in hundreds of dollars } \\
x_{2 i} & =\text { family size }
\end{aligned}
$$
The least squares estimates of the regression parameters were as follows:
$$
b_0=-0.025 \quad b_1=0.052 \quad b_2=1.14
$$
a. Interpret the estimates $b_1$ and $b_2$.
b. Is it possible to provide a meaningful interpretation of the estimate $b_0$ ?

Check back soon!

Problem 9

The following model was fitted to a sample of 25 students using data obtained at the end of their freshman year in college. The aim was to explain students' weight gains:
$$
y_i=\beta_0+\beta_1 x_{1 i}+\beta_2 x_{2 i}+\beta_3 x_{3 i} \varepsilon_i
$$
where
$y_i=$ weight gained, in pounds, during freshman year
$x_{1 i}=$ average number of meals eaten per week
$x_{2 i}=$ average number of hours of exercise per week
$x_{3 i}=$ average number of beers consumed per week
The least squares estimates of the regression parameters were as follows:
$$
b_0=7.35 \quad b_1=0.653 \quad b_2=-1.345 \quad b_3=0.613
$$
a. Interpret the estimates $b_1, b_2$, and $b_3$.
b. Is it possible to provide a meaningful interpretation of the estimate $b_0$ ?

Check back soon!
View

Problem 10

Compute the coefficients $b_1$ and $b_2$ for the regression model
$$
\hat{y}_i=b_0+b_1 x_{1 i}+b_2 x_{2 i}
$$
given the following summary statistics.
a.
$$
\begin{aligned}
& r_{x_1 y}=0.60, r_{x_2 y}=0.70, r_{x_1 x_2}=0.50, \\
& s_{x_1}=200, s_{x_2}=100, s_y=400
\end{aligned}
$$
b.
$$
\begin{aligned}
& r_{x_1 y}=-0.60, r_{x_2 y}=0.70, r_{x_1 x_2}=-0.50, \\
& s_{x_1}=200, s_{x_2}=100, s_y=400
\end{aligned}
$$
c.
$$
\begin{aligned}
& r_{x_1 y}=0.40, r_{x_{2 y}}=0.450, r_{x_1 x_2}=0.80 \\
& s_{x_1}=200, s_{x_2}=100, s_y=400
\end{aligned}
$$
d.
$$
\begin{aligned}
& r_{x_1 y}=0.60, r_{x_{2 y}}=-0.50, r_{x_1 x_2}=-0.60 \\
& s_{x_1}=200, s_{x_2}=100, s_y=400
\end{aligned}
$$

Emily Himsel
Emily Himsel
Numerade Educator
15:32

Problem 11

Consider the following estimated linear regression equations:
$$
Y=a_0+a_1 X_1 \quad Y=b_0+b_1 X_1+b_2 X_2
$$
a. Show in detail the coefficient estimators for $a_1$ and $b_1$ when the correlation between $X_1$ and $X_2$ is equal to 0 .
b. Show in detail the coefficient estimators for $a_1$ and $b_1$ when the correlation between $X_1$ and $X_2$ is equal to 1 .

Paul A.
Paul A.
California State Polytechnic University, Pomona
View

Problem 12

Amalgamated Power, Inc., has asked you to estimate a regression equation to determine the effect of various predictor variables on the demand for electricity sales. You will prepare a series of regression estimates and discuss the results using the quarterly data for electrical sales during the past 17 years in the data file Power Demand.

a. Estimate a regression equation with electricity sales as the dependent variable, using the number of customers and the price as predictor variables. Interpret the coefficients.
b. Estimate a regression equation (electricity sales) using only number of customers as a predictor variable. Interpret the coefficient and compare the result to the result from part a.
c. Estimate a regression equation (electricity sales) using the price and degree days as predictor variables. Interpret the coefficients. Compare the coefficient for price with that obtained in part $a$.
d. Estimate a regression equation (electricity sales) using disposable income and degree days as predictor variables. Interpret the coefficients.

Victor Salazar
Victor Salazar
Numerade Educator
02:44

Problem 13

Transportation Research, Inc., has asked you to prepare some multiple regression equations to estimate the effect of variables on fuel economy. The data for this study are contained in the data file Motors, and the dependent variable is miles per gallon-milpgal-as established by the Department of Transportation certification.
a. Prepare a regression equation that uses vehicle horsepower-horsepower-and vehicle weightweight - as independent variables. Interpret the coefficients.
b. Prepare a second regression equation that adds the number of cylinders - cylinder-as an independent variable to the equation from part a. Interpret the coefficients.
c. Prepare a regression equation that uses number of cylinders and vehicle weight as independent variables. Interpret the coefficients and compare the results with those from parts $a$ and $b$.
$\mathrm{d}$. Prepare a regression equation that uses vehicle horsepower, vehicle weight, and price as predictor variables. Interpret the coefficients.
e. Write a short report that summarizes your results.

Adriano Chikande
Adriano Chikande
Numerade Educator
02:44

Problem 14

Transportation Research, Inc., has asked you to prepare some multiple regression equations to estimate the effect of variables on vehicle horsepower. The data for this study are contained in the data file Motors, and the dependent variable is vehicle horsepower-horsepower-as established by the Department of Transportation certification.
a. Prepare a regression equation that uses vehicle weight-weight-and cubic inches of cylinder displacement-displacement-as predictor variables. Interpret the coefficients.
b. Prepare a regression equation that uses vehicle weight, cylinder displacement, and number of cylinders - cylinder-as predictor variables. Interpret the coefficients and compare the results with those in part a.
c. Prepare a regression equation that uses vehicle weight, cylinder displacement, and miles per gallonmilpgal-as predictor variables. Interpret the coefficients and compare the results with those in part a.
d. Prepare a regression equation that uses vehicle weight, cylinder displacement, miles per gallon, and price as predictor variables. Interpret the coefficients and compare the results with those in part c.
e. Write a short report that presents the results of your analysis of this problem.

Adriano Chikande
Adriano Chikande
Numerade Educator

Problem 15

A regression analysis has produced the following analysis of variance table:
$$
\begin{array}{lrll}
\hline \text { Analysis of Variance } & & & \\
\hline \text { Source } & \text { DF } & \text { SS } & \text { MS } \\
\text { Regression } & 3 & 4,500 & \\
\text { Residual error } & 26 & 500 & \\
\hline
\end{array}
$$
a. Compute $s_e$ and $s_e^2$.
b. Compute SST.
c. Compute $R^2$ and the adjusted coefficient of determination.

Check back soon!

Problem 16

A regression analysis has produced the following analysis of variance table:

$$
\begin{array}{|c|c|c|c|}
\hline {\text { Analysis of Variance }} \\
\hline \text { Source } & \text { DF } & \text { SS } & \text { MS } \\
\hline \text { Regression } & 2 & 7,000 & \\
\hline \text { Residual error } & 29 & 2,500 & \\
\hline
\end{array}
$$

a. Compute $s_e$ and $s_e^2$.
b. Compute SST.
c. Compute $R^2$ and the adjusted coefficient of determination.

Check back soon!

Problem 17

A regression analysis has produced the following analysis of variance table:
$$
\begin{array}{lrll}
\hline {\text { Analysis of Variance }} \\
& \text { DF } & \text { SS } & \text { MS } \\
\hline \text { Source } & 4 & 40,000 & \\
\text { Regression } & 45 & 10,000 & \\
\text { Residual error } & & \\
\hline
\end{array}
$$
a. Compute $s_e$ and $s_e^2$.
b. Compute SST.
c. Compute $R^2$ and the adjusted coefficient of determination.

Check back soon!

Problem 18

A regression analysis has produced the following analysis of variance table:

$$
\begin{array}{lrll}
\hline \text { Analysis of Variance } & & & \\
\hline \text { Source } & \text { DF } & \text { SS } & \text { MS } \\
\text { Regression } & 5 & 80,000 & \\
\text { Residual error } & 200 & 15,000 & \\
\hline
\end{array}
$$

a. Compute $s_e$ and $s_e^2$.
b. Compute SST.
c. Compute $R^2$ and the adjusted coefficient of

Check back soon!

Problem 19

An aircraft company wanted to predict the number of worker-hours necessary to finish the design of a new plane. Relevant explanatory variables were thought to be the plane's top speed, its weight, and the number of parts it had in common with other models built by the company. A sample of 27 of the company's planes was taken, and the following model was estimated:
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\beta_3 x_3+\varepsilon
$$
where
$y=$ design effort, in millions of worker-hours
$x_1=$ plane's top speed, in miles per hour
$x_2=$ plane's weight, in tons
$x_3=$ percentage of parts in common with other models
The estimated regression coefficients were as follows:
$$
b_1=0.661 \quad b_2=0.065 \quad b_3=-0.018
$$
The total sum of squares and regression sum of squares were found to be as follows:
$$
S S T=3.881 \text { and } S S R=3.549
$$
a. Compute and interpret the coefficient of determination.
b. Compute the error sum of squares.
c. Compute the adjusted coefficient of determination.
d. Compute and interpret the coefficient of multiple correlation.

Check back soon!

Problem 20

The following model was fitted to a sample of 30 families in order to explain household milk consumption:
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\varepsilon
$$
where
$y=$ milk consumption, in quarts per week
$x_1=$ weekly income, in hundreds of dollars
$x_2=$ family size
The least squares estimates of the regression parameters were as follows:
$$
b_0=-0.025 \quad b_1=0.052 \quad b_2=1.14
$$
The total sum of squares and regression sum of squares were found to be as follows:
$$
S S T=162.1 \text { and } S S R=88.2
$$
a. Compute and interpret the coefficient of determination.
b. Compute the adjusted coefficient of determination.
c. Compute and interpret the coefficient of multiple correlation.

Check back soon!

Problem 21

The following model was fitted to a sample of 25 students using data obtained at the end of their freshman year in college. The aim was to explain students' weight gains:
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\beta_3 x_3+\varepsilon
$$
where
$y=$ weight gained, in pounds, during freshman year
$x_1=$ average number of meals eaten per week
$x_2=$ average number of hours of exercise per week
$x_3=$ average number of beers consumed per week
The least squares estimates of the regression parameters were as follows:
$$
b_0=7.35 \quad b_1=0.653 \quad b_2=-1.345 \quad b_3=0.613
$$
The regression sum of squares and error sum of squares were found to be as follows:
$$
S S R=79.2 \text { and } S S E=45.9
$$
a. Compute and interpret the coefficient of determination.
b. Compute the adjusted coefficient of determination.
c. Compute and interpret the coefficient of multiple correlation.

Check back soon!
02:15

Problem 22

Refer to the savings and loan association data given in Table 12.1.
a. Estimate, by least squares, the regression of profit margin on number of offices.
b. Estimate, by least squares, the regression of net revenues on number of offices.
c. Estimate, by least squares, the regression of profit margin on net revenues.
d. Estimate, by least squares, the regression of number of offices on net revenues.

Eric Mockensturm
Eric Mockensturm
Numerade Educator

Problem 23

The following are results from a regression model analysis:
$$
\begin{gathered}
\hat{y}=1.50+\underset{(2.1)}{4.8 x_1}+\underset{(3.7)}{6.9 x_2}-\underset{(2.8)}{7.2 x_3} \\
R^2=0.71 \quad n=24
\end{gathered}
$$
The numbers below the coefficient estimates are the sample standard errors of the coefficient estimates.
a. Compute two-sided $95 \%$ confidence intervals for the three regression slope coefficients.
b. For each of the slope coefficients, test the hypothesis

Check back soon!

Problem 24

The following are results from a regression model analysis:
$$
\begin{aligned}
& \hat{y}=2.50+6.8 x_1+6.9 x_2-7.2 x_3 \\
& \text { (3.1) } \\
& \text { (3.7) } \\
& \text { (3.2) } \\
& R^2=0.85 \\
& n=34 \\
&
\end{aligned}
$$
The numbers below the coefficient estimates are the estimated coefficient standard errors.
a. Compute two-sided $95 \%$ confidence intervals for the three regression slope coefficients.
b. For each of the slope coefficients test the hypothesis
$$
H_0: \beta_j=0
$$

Check back soon!
View

Problem 25

The following are results from a regression model analysis:
$$
\begin{array}{cc}
\hat{y}=-101.50+\underset{(12.1]}{34.8 x_1}+\underset{(23.7)}{56.9 x_2}-\underset{(32.8)}{57.2 x_3} \\
R^2=0.71 & n=65
\end{array}
$$
The numbers in parentheses under the coefficients are the estimated coefficient standard errors.
a. Compute two-sided $95 \%$ confidence intervals for the three regression slope coefficients.
b. For each of the slope coefficients test the hypothesis
$$
H_0: \beta_j=0
$$

Rashmi Sinha
Rashmi Sinha
Numerade Educator
View

Problem 26

The following are results from a regression model analysis:
$$
\begin{array}{cc}
\hat{y}=-101.50+\underset{(12.1]}{34.8 x_1}+\underset{(23.7)}{56.9 x_2}-\underset{(32.8)}{57.2 x_3} \\
R^2=0.71 & n=65
\end{array}
$$
The numbers in parentheses under the coefficients are the estimated coefficient standard errors.
a. Compute two-sided $95 \%$ confidence intervals for the three regression slope coefficients.
b. For each of the slope coefficients test the hypothesis
$$
H_0: \beta_j=0
$$

Rashmi Sinha
Rashmi Sinha
Numerade Educator
View

Problem 27

An aircraft company wanted to predict the number of worker-hours necessary to finish the design of a new plane. Relevant explanatory variables were thought to be the plane's top speed, its weight, and the number of parts it had in common with other models built by the company. A sample of 27 of the company's planes was taken, and the following model was estimated:
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\beta_3 x_3+\varepsilon
$$
where
$y=$ design effort, in millions of worker-hours
$x_1=$ plane's top speed, in miles per hour
$x_2=$ plane's weight, in tons
$x_3=$ percentage of parts in common with other models

The estimated regression coefficients were as follows:
$$
b_1=0.661 \quad b_2=0.065 \quad b_3=-0.018
$$
The estimated standard errors were as follows:
$$
s_{b_1}=0.099 \quad \mathrm{~s}_{b_2}=0.032 \quad \mathrm{~s}_{b_3}=0.0023
$$
a. Find $90 \%$ and $95 \%$ confidence intervals for $\beta_1$.
b. Find $95 \%$ and $99 \%$ confidence intervals for $\beta_2$.
c. Test against a two-sided alternative the null hypothesis that, all else being equal, the plane's weight has no linear influence on its design effort.
d. The error sum of squares for this regression was 0.332 . Using the same data, a simple linear regression of design effort on the percentage of common parts was fitted, yielding an error sum of squares of 3.311. Test, at the $1 \%$ level, the null hypothesis that, taken together, the variable's top speed and weight contribute nothing in a linear sense to explaining the changes in the variable, design effort, given that the variable percentage of common parts is also used as an explanatory variable.

Shu Naito
Shu Naito
Numerade Educator

Problem 28

The following model was fitted to a sample of 30 families in order to explain household milk consumption:
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\varepsilon
$$
where
$$
\begin{aligned}
y & =\text { milk consumption, in quarts per week } \\
x_1 & =\text { weekly income, in hundreds of dollars } \\
x_2 & =\text { family size }
\end{aligned}
$$
The least squares estimates of the regression parameters were as follows:
$$
b_0=-0.025 \quad b_1=0.052 \quad b_2=1.14
$$
The estimated standard errors were as follows:
$$
s_{h_1}=0.023 \quad \mathrm{~s}_{k_2}=0.35
$$
a. Test, against the appropriate one-sided alternative, the null hypothesis that, for fixed family size, milk consumption does not depend linearly on income.
b. Find $90 \%, 95 \%$, and $99 \%$ confidence intervals for $\beta_2$.

Check back soon!

Problem 29

The following model was fitted to a sample of 25 students using data obtained at the end of their freshman year in college. The aim was to explain students' weight gains:
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\beta_3 x_3+\varepsilon
$$

where
$y=$ weight gained, in pounds, during freshman year
$x_1=$ average number of meals eaten per week
$x_2=$ average number of hours of exercise per week
$x_3=$ average number of beers consumed per week
The least squares estimates of the regression parameters were as follows:
$$
b_0=7.35 \quad b_1=0.653 \quad b_2=-1.345 \quad b_3=0.613
$$
The estimated standard errors were as follows:
$$
s_{b_1}=0.189 \quad s_{b_2}=0.565 \quad s_{b_3}=0.243
$$
a. Test, against the appropriate one-sided alternative, the null hypothesis that, all else being equal, hours of exercise do not linearly influence weight gain.
b. Test, against the appropriate one-sided alternative, the null hypothesis that, all else being equal, beer consumption does not linearly influence weight gain.
c. Find $90 \%, 95 \%$, and $99 \%$ confidence intervals for $\beta_1$.

Check back soon!
01:47

Problem 30

Refer to the data of Example 12.6.
a. Test, against a two-sided alternative, the null hypothesis that, all else being equal, median per capita personal income has no influence on the effective property tax rate.
b. Test the null hypothesis that, taken together, the three independent variables do not linearly influence the effective property tax rate.

Karen Song
Karen Song
Numerade Educator
03:51

Problem 31

Refer to the data of Example 12.7 with the data file Citydatr.
a. Find $95 \%$ and $99 \%$ confidence intervals for the expected change in the market price for houses resulting from a one-unit increase in the mean number of rooms when the values of all other independent variables remain unchanged.
b. Test the null hypothesis that, all else being equal, mean household income does not influence the market price against the alternative that the higher the mean household income, the higher the market price.

James Kiss
James Kiss
Numerade Educator

Problem 32

In a study of revenue generated by national lotteries, the following regression equation was fitted to data from 29 countries with lotteries:
$$
\begin{aligned}
& y=-31.323+\underset{(0.00755)}{0.4045 x_1}+\underset{(0.3107)}{0.8772 x_2}-\underset{(263.83)}{365.01 x_3}-\underset{(3.4520)}{9.9298 x_4} \\
& R^2=.51
\end{aligned}
$$
where
$$
\begin{aligned}
y= & \text { dollars of net revenue per capita per year gen- } \\
& \text { erated by the lottery } \\
x_1= & \text { mean per capita personal income of the } \\
& \text { country } \\
x_2= & \text { number of hotel, motel, inn, and resort rooms } \\
& \text { per thousand persons in the country } \\
x_3= & \text { spendable revenue per capita per year gener- } \\
& \text { ated by pari-mutuel betting, racing, and other } \\
& \text { legalized gambling } \\
x_4= & \text { percentage of the nation's border contiguous } \\
& \text { with a state or states with a lottery }
\end{aligned}
$$
The numbers in parentheses under the coefficients are the estimated coefficient standard errors.
a. Interpret the estimated coefficient on $x_1$.
b. Find and interpret a $95 \%$ confidence interval for the coefficient on $x_2$ in the population regression.
c. Test the null hypothesis that the coefficient on $x_3$ in the population regression is 0 against the alternative that this coefficient is negative. Interpret your findings.

Check back soon!

Problem 33

A study was conducted to determine whether certain features could be used to explain variability in the prices of furnaces. For a sample of 19 furnaces, the following regression was estimated:
$$
y=-68.236+\underset{(0.005)}{0.0023 x_1}+\underset{(8.992)}{19.729 x_2}+\underset{(3.082)}{7.653 x_3} \quad R^2=0.84
$$
where
$$
\begin{aligned}
y & =\text { price, in dollars } \\
x_1 & =\text { rating of furnace, in BTU per hour } \\
x_2 & =\text { energy efficiency ratio } \\
x_3 & =\text { number of settings }
\end{aligned}
$$
The numbers in parentheses under the coefficients are the estimated coefficient standard errors.
a. Find a $95 \%$ confidence interval for the expected increase in price resulting from an additional setting when the values of the rating and the energy efficiency ratio remain fixed.
b. Test the null hypothesis that, all else being equal, the energy efficiency ratio of furnaces does not affect their price against the alternative that the higher the energy efficiency ratio, the higher the price.

Check back soon!
07:03

Problem 34

In a study of differences in levels of community demand for firefighters, the following sample regression was obtained, based on data from 39 towns in Maryland:
$$
\begin{aligned}
y= & -\underset{(0.00010)}{0.00232}-\underset{(0.000018)}{0.00024 x_1}-\underset{(0.00012)}{0.00002 x_2}+\underset{(0.00005)}{0.00034 x_3} \\
& +\underset{(0.77951)}{0.48122 x_4}+\underset{(0.01172)}{0.04950 x_5}-\underset{(0.06306)}{0.00010 x_6}+\underset{\left(0.00645 x_7\right.}{0.0 .3572} \\
R^2= & 0.357
\end{aligned}
$$
(0.00005)
(0.00306)
where
$y=$ number of full-time firefighters per capita
$x_1=$ maximum base salary of firefighters, in thousands of dollars
$x_2=$ percentage of population
$x_3=$ estimated per capita income, in thousands of dollars
$x_4=$ population density
$x_5=$ amount of intergovernmental grants per capita, in thousands of dollars
$x_6=$ number of miles from the regional city
$x_7=$ percentage of the population that is male and between 12 and 21 years of age

The numbers in parentheses under the coefficients are the estimated coefficient standard errors.
a. Find and interpret a $99 \%$ confidence interval for $\beta_5$.
b. Test, against a two-sided alternative, the null hypothesis that $\beta_4$ is 0 , and interpret your result.
c. Test, against a two-sided alternative, the null hypothesis that $\beta_7$ is 0 , and interpret your result.

Jon Southam
Jon Southam
Numerade Educator

Problem 36

An aircraft company wanted to predict the number of worker-hours necessary to finish the design of a new plane. Relevant explanatory variables were thought to be the plane's top speed, its weight, and the number of parts it had in common with other models built by the company. A sample of 27 of the company's planes was taken, and the following model was estimated:
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\beta_3 x_3+\varepsilon
$$
where
$y=$ design effort, in millions of worker-hours
$x_1=$ plane's top speed, in miles per hour
$x_2=$ plane's weight, in tons
$x_3=$ percentage of parts in common with other models

The estimated regression coefficients were as follows:
$$
b_1=0.661 \quad b_2=0.065 \quad b_3=-0.018
$$
The total sum of squares and regression sum of squares were found to be as follows:
$$
S S T=3.881 \text { and } S S R=3.549
$$
a. Test the null hypothesis:
$$
H_0: \beta_1=\beta_2=\beta_3=0
$$
b. Set out the analysis of variance table.

Check back soon!
01:41

Problem 37

In a study of the influence of financial institutions on bond interest rates in Germany, quarterly data over a period of 12 years were analyzed. The postulated model was
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\varepsilon
$$
where
$y=$ change over the quarter in the bond interest rates
$x_1=$ change over the quarter in bond purchases by financial institutions
$x_2=$ change over the quarter in bond sales by financial institutions

The estimated partial regression coefficients were as follows:
$$
b_1=0.057 \quad b_2=-0.065
$$
The corrected coefficient of determination was found to be $R^2=0.463$. Test the null hypothesis:
$$
H_0: \beta_1=\beta_2=0
$$

Sheryl Ezze
Sheryl Ezze
Numerade Educator

Problem 38

The following model was fitted to a sample of 30 families in order to explain household milk consumption:
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\varepsilon
$$
where
$$
\begin{aligned}
y & =\text { milk consumption, in quarts per week } \\
x_1 & =\text { weekly income, in hundreds of dollars } \\
x_2 & =\text { family size }
\end{aligned}
$$
The least squares estimates of the regression parameters were as follows:
$$
b_0=-0.025 \quad b_1=0.052 \quad b_2=1.14
$$
The estimated standard errors were as follows:
$$
s_{b_1}=0.023 \quad s_{b_2}=0.35
$$
The total sum of squares and regression sum of squares were found to be as follows:
$$
S S T=162.1 \text { and } S S R=88.2
$$
a. Test the null hypothesis:
$$
H_0: \beta_1=\beta_2=0
$$
b. Set out the analysis of variance table.

Check back soon!

Problem 39

The following model was fitted to a sample of 25 students using data obtained at the end of their freshman year in college. The aim was to explain students' weight gains:
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\beta_3 x_3+\varepsilon
$$
where
$y=$ weight gained, in pounds, during freshman year
$x_1=$ average number of meals eaten per week
$x_2=$ average number of hours of exercise per week
$x_3=$ average number of beers consumed per week
The least squares estimates of the regression parameters were as follows:
$$
b_0=7.35 \quad b_1=0.653 \quad b_2=-1.345 \quad b_3=0.613
$$
The estimated standard errors were as follows:
$$
s_{b_1}=0.189 \quad s_{b_2}=0.565 \quad s_{b_3}=0.243
$$
The regression sum of squares and error sum of squares were found to be as follows:
$$
S S R=79.2 \text { and } S S E=45.9
$$
a. Test the null hypothesis:
$$
H_0: \beta_1=\beta_2=\beta_3=0
$$
b. Set out the analysis of variance table.

Check back soon!

Problem 40

A dependent variable is regressed on $K$ independent variables, using $n$ sets of sample observations. We denote SSE as the error sum of squares and $R^2$ as the coefficient of determination for this estimated regression. We want to test the null hypothesis that $K_1$ of these independent variables, taken together, do not linearly affect the dependent variable, given that the other $\left(K-K_1\right)$ independent variables are also to be used. Suppose that the regression is reestimated with the $K_1$ independent variables of interest excluded. Let $S S E^*$ denote the error sum of squares and $R^{* 2}$, the coefficient of determination for this regression. Show that the statistic for testing our null hypothesis, introduced in Section 12.5, can be expressed as follows:
$$
\frac{\left(S S E^*-S S E\right) / K_1}{S S E /(n-K-1)}=\frac{R^2-R^{* 2}}{1-R^2} \cdot \frac{n-K-1}{K_1}
$$

Check back soon!

Problem 41

The following model was fitted to a sample of 30 families in order to explain household milk consumption:
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\varepsilon
$$

where
$$
\begin{aligned}
& y=\text { milk consumption, in quarts per week } \\
& x_1=\text { weekly income, in hundreds of dollars } \\
& x_2=\text { family size }
\end{aligned}
$$
The least squares estimates of the regression parameters were as follows:
$$
b_0=-0.025 \quad b_1=0.052 \quad b_2=1.14
$$
The total sum of squares and regression sum of squares were found to be as follows:
$$
S S T=162.1 \text { and } S S R=88.2
$$
A third independent variable-number of preschool children in the household-was added to the regression model. The sum of squared errors when this augmented model was estimated by least squares was found to be 83.7 . Test the null hypothesis that, all other things being equal, the number of preschool children in the household does not linearly affect milk consumption.

Check back soon!

Problem 42

Suppose that a dependent variable is related to $K$ independent variables through a multiple regression model. Let $R^2$ denote the coefficient of determination and $\bar{R}^2$, the corrected coefficient. Suppose that $n$ sets of observations are used to fit the regression.
a. Show that
$$
\bar{R}^2=\frac{(n-1) R^2-K}{n-K-1}
$$
b. Show that
$$
R^2=\frac{(n-K-1) \bar{R}^2+K}{n-1}
$$
c. Show that the statistic for testing the null hypothesis that all the regression coefficients are 0 can be written as
$$
\frac{S S R / K}{S S E /(n-K-1)}=\frac{n-K-1}{K} \cdot \frac{\bar{R}^2+A}{1-\bar{R}^2}
$$
where
$$
A=\frac{K}{n-K-1}
$$

Check back soon!
05:16

Problem 43

Given the estimated multiple regression equation
$$
\hat{y}=6+5 x_1+4 x_2+7 x_3+8 x_4
$$
what is the predicted value of $Y$ in each case?
a. $x_1=10, x_2=23, x_3=9$, and $x_4=12$
b. $x_1=23, x_2=18, x_3=10$, and $x_4=11$
c. $x_1=10, x_2=23, x_3=9$, and $x_4=12$
d. $x_1=-10, x_2=13, x_3=-8$, and $x_4=-16$

Sneha Ravi
Sneha Ravi
Numerade Educator

Problem 44

The following model was fitted to a sample of 25 students using data obtained at the end of their freshman year in college. The aim was to explain students' weight gains:
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\beta_3 x_3+\varepsilon
$$
where
$y=$ weight gained, in pounds, during freshman year
$x_1=$ average number of meals eaten per week
$x_2=$ average number of hours of exercise per week
$x_3=$ average number of beers consumed per week
The least squares estimates of the regression parameters were as follows:
$$
b_0=7.35 \quad b_1=0.653 \quad b_2=-1.345 \quad b_3=0.613
$$
Predict the weight gain for a freshman who eats an average of 20 meals per week, exercises an average of 10 hours per week, and consumes an average of 6 beers per week.

Check back soon!

Problem 45

The following model was fitted to a sample of 30 families in order to explain household milk consumption:
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\varepsilon
$$
where
$y=$ milk consumption, in quarts per week
$x_1=$ weekly income, in hundreds of dollars
$x_2=$ family size
The least squares estimates of the regression parameters were as follows:
$$
b_0=-0.025 \quad b_1=0.052 \quad b_2=1.14
$$
Predict the weekly milk consumption of a family of four with an income of $\$ 600$ per week.

Check back soon!
View

Problem 47

A real estate agent hypothesizes that in her town the selling price of a house in dollars (y) depends on its size in square feet of floor space $\left(x_1\right)$, the lot size in square feet $\left(x_2\right)$, the number of bedrooms $\left(x_3\right)$, and the number of bathrooms $\left(x_4\right)$. For a random sample of 20 house sales, the following least squares estimated model was obtained:

$$
\begin{aligned}
& \hat{y}=1998.5+\underset{(2.5543)}{22.352 x_1}+\underset{(1.4492)}{1.4686 x_2}+\underset{(1820.8)}{6767.3 x_3}+\underset{(1996.2)}{2701.1 x_4} \\
& R^2=0.9843
\end{aligned}
$$
The numbers in parentheses under the coefficients are the estimated coefficient standard errors.
a. Interpret in the context of this model the estimated coefficient on $x_2$.
b. Interpret the coefficient of determination.
c. Assuming that the model is correctly specified, test, at the $5 \%$ level against the appropriate onesided alternative, the null hypothesis that, all else being equal, selling price does not depend on number of bathrooms.
d. Estimate the selling price of a house with 1,250 square feet of floor space, a lot of 4,700 square feet, 3 bedrooms, and 1 bathroom.

Shu Naito
Shu Naito
Numerade Educator
02:44

Problem 48

Transportation Research, Inc., has asked you to prepare a multiple regression equation to estimate the effect of variables on fuel economy. The data for this study are contained in the data file Motors, and the dependent variable is miles per gallonmilpgal-as established by the Department of Transportation certification.
a. Prepare a regression equation that uses vehicle horsepower-horsepower-and vehicle weightweight - as independent variables. Determine the predicted value, the confidence interval of the prediction, and the prediction interval when the horsepower is 140 and the vehicle weight is 3,000 pounds.
b. Prepare a second regression equation that adds the number of cylinders-cylinder-as an independent variable to the equation from part a. Determine the predicted value, the confidence interval of the prediction, and the prediction interval when the horsepower is 140 , the number of cylinders is 6 and the vehicle weight is 3,000 pounds.

Adriano Chikande
Adriano Chikande
Numerade Educator
View

Problem 49

Consider the following two equations estimated using the procedures developed in this section:
i. $y_i=4 x^{1.5}$
ii. $y_i=1+2 x_i+2 x_i^2$
Compute values of $y_i$ when $x_i=1,2,4,6,8,10$.

Thomas Tamanaha
Thomas Tamanaha
Numerade Educator
View

Problem 50

Consider the following two equations estimated using the procedures developed in this section:
i. $y_i=4 x^{1.8}$
ii. $y_i=1+2 x_i+2 x_i^2$
Compute values of $y_i$ when $x_i=1,2,4,6,8,10$.

Thomas Tamanaha
Thomas Tamanaha
Numerade Educator
View

Problem 51

Consider the following two equations estimated using the procedures developed in this section:
i. $y_i=4 x^{1.5}$
ii. $y_i=1+2 x_i+1.7 x_i^2$
Compute values of $y_i$ when $x_i=1,2,4,6,8,10$.

Thomas Tamanaha
Thomas Tamanaha
Numerade Educator

Problem 52

Consider the following two equations estimated using the procedures developed in this section.
i. $y_i=3 x^{1.2}$
ii. $y_i=1+5 x_i-1.5 x_i^2$
Compute values of $y_i$ when $x_i=1,2,4,6,8,10$.

Check back soon!
02:24

Problem 53

Describe an example from your experience in which a quadratic model would be better than a linear model.

Carson Merrill
Carson Merrill
Numerade Educator
View

Problem 54

John Swanson, president of Market Research Inc., has asked you to estimate the coefficients of the model
$$
Y=\beta_0+\beta_1 X_1+\beta_2 X_1^2+\beta_3 X_2
$$
where $Y$ is the expected sales of office supplies for a large retail distributor of office supplies, $X_1$ is the total disposable income of residents within 5 miles of the store, and $X_2$ is the total number of persons employed in information-based businesses within 5 miles of the store. Recent work by a national consulting firm has concluded that the coefficients in the model must have the following restriction:
$$
\beta_1+\beta_2=2
$$
Describe how you would estimate the model coefficients using least squares.

Rashmi Sinha
Rashmi Sinha
Numerade Educator
03:05

Problem 55

In a study of the determinants of household expenditures on vacation travel, data were obtained from a sample of 2,246 households (Hagermann 1981). The model estimated was
$$
\begin{aligned}
& \log y=-4.054+1.1556 \log x_1-0.4408 \log x_2 \\
& (0.0546) \\
& R^2=.168 \\
&
\end{aligned}
$$
$(0.0490)$
where
$y=$ expenditure on vacation travel
$x_1=$ total annual consumption expenditure
$x_2=$ number of members in household
The numbers in parentheses under the coefficients are the estimated coefficient standard errors.
a. Interpret the estimated regression coefficients.
b. Interpret the coefficient of determination.
c. All else being equal, find a $95 \%$ confidence interval for the percentage increase in expenditures on vacation travel resulting from a $1 \%$ increase in total annual consumption expenditures.
d. Assuming that the model is correctly specified, test, at the $1 \%$ significance level, the null hypothesis that, all else being equal, the number of members in a household does not affect expenditures on vacation travel against the alternative that the greater the number of household members, the lower the vacation travel expenditures.

James Kiss
James Kiss
Numerade Educator
02:12

Problem 56

The following model was estimated for a sample of 322 supermarkets in large metropolitan areas (Macdonald and Nelson 1991):
$$
\begin{aligned}
\log (y) & =2.921+\underset{(0.077)}{0.680} \log (x) \\
R^2 & =0.19
\end{aligned}
$$
where
$y=$ store size
$x=$ median income in zip-code area in which store is located

The number in parentheses under the coefficient is the estimated coefficient standard error.
a. Interpret the estimated coefficient on $\log x$.
b. Test the null hypothesis that income has no impact on store size against the alternative that higher income tends to be associated with larger store size.

Adriano Chikande
Adriano Chikande
Numerade Educator
01:42

Problem 57

An agricultural economist believes that the amount of beef consumed (y) in tons in a year in the United States depends on the price of beef $\left(x_1\right)$ in dollars per pound, the price of pork $\left(x_2\right)$ in dollars per pound, the price of chicken $\left(x_3\right)$ in dollars per pound, and the income per household $\left(x_4\right)$ in thousands of dollars. The following sample regression was obtained through least squares, using 30 annual observations:
$\log y=-0.024-0.529 \log x_1+0.217 \log x_2+0.193 \log x_3$
(0.168)
(0.103)
(0.106)
$+0.416 \log x_4 \quad R^2=0.683$
$(.163)$
The numbers in parentheses under the coefficients are the estimated coefficient standard errors.
a. Interpret the coefficient on $\log x_1$.
b. Interpret the coefficient on $\log x_2$.
c. Test, at the $1 \%$ significance level, the null hypothesis that the coefficient on $\log x_4$ in the population regression is 0 against the alternative that it is positive.
d. Test the null hypothesis that the four variables $\left(\log x_1, \log x_2, \log x_3, \log x_4\right)$ do not, as a set, have any linear influence on $\log y$.
e. The economist is also concerned that, over the years, the increasing awareness of the effects of heavy red-meat consumption on health may have influenced the demand for beef. If this is indeed the case, how would this influence your view of the original estimated regression?

Adriano Chikande
Adriano Chikande
Numerade Educator

Problem 58

You have been asked to develop an exponential production function-Cobb-Douglas form-that will predict the number of microprocessors produced by a manufacturer, $Y$, as a function of the units of capital, $X_1$; the units of labor, $X_2$; and the number of computer science staff involved in basic research, $X_3$. Specify the model form and then carefully and completely indicate how you would estimate the coefficients. Do this first using an unrestricted model and then a second time including the restriction that the coefficients of the three variables should sum to 1 .

Check back soon!
04:14

Problem 59

Consider the following nonlinear model with multiplicative errors:
$$
\begin{aligned}
Y= & \beta_0 X_1^{\beta_1} X_2^{\beta_2} X_3^{\beta_3} X_4^{\beta_4} \varepsilon \\
& \beta_1+\beta_2=1 \\
& \beta_3+\beta_4=1
\end{aligned}
$$
a. Show how you would obtain the coefficient estimates. Coefficient restrictions must be satisfied. Show all your work and explain what you are doing.
b. What is the constant elasticity for $Y$ versus $X_4$ ? Show all your work.

Victor Salazar
Victor Salazar
Numerade Educator

Problem 60

Angelica Chandra, president of Benefits Research, Inc., has asked you to study the salary structure of her firm. Benefits Research provides consulting and management for employee health care and retirement programs. Its clients are mid- to large-sized firms. As a first step you are asked to estimate a regression model that estimates expected salary as a function of years of experience in the firm. You are to consider linear, quadratic, and cubic models and determine which one would be most suitable. Estimate appropriate regression models and write a short report that recommends the best model. Use the data contained in the file Benefits Research.

Check back soon!
02:11

Problem 61

The data file German Imports shows German real imports (y), real private consumption $\left(x_1\right)$, and real exchange rate $\left(x_2\right)$, in terms of U.S. dollars per mark, over a period of 22 years. Estimate the model
$$
\log y_t=\beta_0+\beta_1 \log x_{1 t}+\beta_2 \log x_{2 t}+\varepsilon_i
$$
and write a report on your findings.

Adriano Chikande
Adriano Chikande
Numerade Educator
02:07

Problem 62

What is the model constant when the dummy variable equals 1 in the following equations, where $x_1$ is a continuous variable and $x_2$ is a dummy variable with a value of 0 or 1 ?
a. $\hat{y}=4+8 x_1+3 x_2$
b. $\hat{y}=7+6 x_1+5 x_2$
c. $\hat{y}=4+8 x_1+3 x_2+4 x_1 x_2$

Jason Orozco
Jason Orozco
Numerade Educator
02:46

Problem 63

What are the model constant and the slope coefficient of $x_1$ when the dummy variable equals 1 in the following equations, where $x_1$ is a continuous variable and $x_2$ is a dummy variable with a value of 0 or 1 ?
a. $\hat{y}=4+9 x_1+1.78 x_2+3.09 x_1 x_2$
b. $\hat{y}=-3+7 x_1+4.15 x_2+2.51 x_1 x_2$
c. $\hat{y}=10+5 x_1+3.67 x_2+3.98 x_1 x_2$

Narayan Hari
Narayan Hari
Numerade Educator
03:07

Problem 64

The following model was fitted to observations from 1972 to 1979 in an attempt to explain oil-pricing behavior:
$$
\hat{y}=\underset{(0.029)}{37 x_1}+\underset{(0.50)}{5.22 x_2}
$$
where
$\hat{y}=$ difference between price in the current year and price in the previous year, in dollars per barrel
$x_1=$ difference between spot price in the current year and spot price in the previous year
$x_2=$ dummy variable taking the value 1 in 1974 and 0 otherwise to represent the specific effect of the oil embargo of that year

The numbers in parentheses under the coefficients are the estimated coefficient standard errors.

Interpret verbally and graphically the estimated coefficient on the dummy variable.

Chengyu Li
Chengyu Li
Numerade Educator
View

Problem 65

The following model was fitted to explain the selling prices of condominiums in a sample of 815 sales:
$$
\begin{aligned}
\hat{y}= & \underset{(0.91)}{-1264}+\underset{(515)}{48.18 x_1}+\underset{(488)}{3382 x_2}-\underset{(947)}{1859 x_3}+\underset{(768)}{3219 x_4} \\
& +2005 x_5 \quad \bar{R}^2=0.86
\end{aligned}
$$
(768)
where
$\hat{y}=$ selling price of condo, in dollars
$x_1=$ square footage of living area
$x_2=$ size of garage, in number of cars
$x_3=$ age of condo, in years
$x_4=$ dummy variable taking the value 1 if the condo has a fireplace and 0 otherwise
$x_5=$ dummy variable taking the value 1 if the condo has hardwood floors and 0 if it has vinyl floors
a. Interpret the estimated coefficient of $x_4$.
b. Interpret the estimated coefficient of $x_5$.
c. Find a $95 \%$ confidence interval for the impact of a fireplace on selling price, all other things being equal.
d. Test the null hypothesis that type of flooring has no impact on selling price against the alternative that, all other things equal, condos with hardwood floors have a higher selling price than those with vinyl flooring.

Shu Naito
Shu Naito
Numerade Educator
02:35

Problem 66

The following model was fitted to data on 32 insurance companies:
$$
\hat{y}=7.62-\underset{(0.008)}{0.16 x_1}+\underset{(0.496)}{1.23 x_2} \quad R^2=0.37
$$
where
$$
\begin{aligned}
\hat{y}= & \text { price-earnings ratio } \\
x_1= & \text { size of insurance company assets, in billions } \\
& \text { of dollars } \\
x_2= & \text { dummy variable taking the value } 1 \text { for regional } \\
& \text { companies and } 0 \text { for national companies }
\end{aligned}
$$
The numbers in parentheses under the coefficients are the estimated coefficient standard errors.
a. Interpret the estimated coefficient on the dummy variable.
b. Test against a two-sided alternative. the null hypothesis that the true coefficient on the dummy variable is 0 .
c. Test, at the $5 \%$ level, the null hypothesis $\beta_1=\beta_2=0$, and interpret your result.

Nick Johnson
Nick Johnson
Numerade Educator

Problem 67

A business school dean wanted to assess the importance of factors that might help in predicting success in law school. For a random sample of 50 students, data were obtained when students graduated from law school, and the following model was fitted:
$$
y=\alpha+\beta_1 x_1+\beta_2 x_2+\beta_3 x_3+\varepsilon
$$
where
$$
\begin{aligned}
y= & \text { score reflecting overall performance while in } \\
& \text { law school } \\
x_1= & \text { undergraduate grade point average } \\
x_2= & \text { score on GMAT } \\
x_3= & \text { dummy variable taking the value } 1 \text { if a stu- } \\
& \text { dent's letters of recommendation are unusually } \\
& \text { strong and } 0 \text { otherwise }
\end{aligned}
$$
Use the portion of the computer output from the estimated regression shown here to write a report summarizing the findings of this study.
$$
\begin{array}{lrrrcc}
\hline \text { Source } & \text { DF } & \begin{array}{c}
\text { Sum of } \\
\text { Squares }
\end{array} & \begin{array}{c}
\text { Mean } \\
\text { Square }
\end{array} & \begin{array}{c}
F \\
\text { Value }
\end{array} & \begin{array}{c}
R- \\
\text { Square }
\end{array} \\
\hline \text { Model } & 3 & 641.04 & 212.68 & 8.48 & 0.356 \\
\text { Error } & 46 & 1,159.66 & 25.21 & & \\
\text { Total } & 49 & 1,800.70 & & & \\
\hline
\end{array}
$$
$$
\begin{array}{cccc}
\hline \text { Parameter } & \text { Estimate } & \begin{array}{c}
t \text { for } H_0: \\
\beta_j=0
\end{array} & \begin{array}{c}
\text { Std. Error of } \\
\text { Estimate }
\end{array} \\
\hline \text { Intercept } & 6.512 & & \\
\text { X1 } & 3.502 & 1.45 & 2.419 \\
\text { X2 } & 0.491 & 4.59 & 0.107 \\
\text { X3 } & 10.327 & 2.45 & 4.213 \\
\hline
\end{array}
$$

Check back soon!

Problem 68

The following model was fitted to data on 50 states:
where
$$
\begin{aligned}
& \hat{y}=\text { annual salary of the attorney general of the } \\
& \text { state } \\
& x_1=\text { average annual salary of lawyers, in thou- } \\
& \text { sands of dollars } \\
& x_2=\text { number of bills enacted in previous legisla- } \\
& \text { tive session } \\
& x_3=\text { number of due process reviews by state } \\
& \text { courts that resulted in overturn of legislation } \\
& \text { in previous } 40 \text { years } \\
& x_4=\text { length of term of the attorney general of the } \\
& \text { state } \\
& x_5=\text { dummy variable taking value } 1 \text { if justices of } \\
& \text { the state supreme court can be removed from } \\
&
\end{aligned}
$$
office by the governor, judicial review board, or majority vote of the supreme court and 0 otherwise
$x_6=$ dummy variable taking value 1 if supreme court justices are elected on partisan ballots and 0 otherwise
The numbers in parentheses under the coefficients are the estimated coefficient standard errors.
a. Interpret the estimated coefficient on the dummy variable $x_5$.
b. Interpret the estimated coefficient on the dummy variable $x_6$.
c. Test, at the $5 \%$ level, the null hypothesis that the true coefficient on the dummy variable $x_5$ is 0 against the alternative that it is positive.
d. Test, at the $5 \%$ level, the null hypothesis that the true coefficient on the dummy variable $x_6$ is 0 against the alternative that it is negative.
e. Find and interpret a $95 \%$ confidence level for the parameter $\beta_1$.

Check back soon!
01:42

Problem 69

A consulting group offers courses in financial management for executives. At the end of these courses participants are asked to provide overall ratings of the value of the course. For a sample of 25 courses, the following regression was estimated by least squares:
$$
\hat{y}=42.97+\underset{(0.29)}{0.38 x_1}+\underset{(0.21)}{0.52 x_2} \underset{(0.11)}{0.08 x_3}+\underset{(0.359)}{6.21 x_4} \quad R^2=0.569
$$
where
$\hat{y}=$ average rating by participants of the course
$x_1=$ percentage of course time spent in group discussion sessions
$x_2=$ money, in dollars, per course member spent on preparing course material
$x_3=$ money, in dollars, per course member spent on food and drinks
$x_4=$ dummy variable taking the value 1 if a visiting guest lecturer is brought in and 0 otherwise
The numbers in parentheses under the coefficients are the estimated coefficient standard errors.
a. Interpret the estimated coefficient on $x_4$.
b. Test, against the alternative that it is positive, the null hypothesis that the true coefficient on $x_4$ is 0 .
c. Interpret the coefficient of determination, and use it to test the null hypothesis that, taken as a group, the four independent variables do not linearly influence the dependent variable.
d. Find and interpret a $95 \%$ confidence interval for $\beta_2$.

Adriano Chikande
Adriano Chikande
Numerade Educator

Problem 70

A regression model was estimated to compare performance of students taking a business statistics courseeither as a standard 14-week course or as an intensive 3-week course. The following model was estimated from observations of 350 students (Van Scyoc and Gleason 1993):
$$
\begin{aligned}
& \hat{y}=-.7052+1.4170 x_1+2.1624 x_2+.8680 x_3+1.0845 x_4 \\
& (0.4568) \\
& (0.3287) \\
& +0.4694 x_5+0.0038 x_6+0.0484 x_7 \quad R^2=0.344 \\
&
\end{aligned}
$$
(.4393)
$(0.3766)$
(0.0628)
(0.0094)
(0.0776)
where
$\hat{y}=$ score on a standardized test of understanding of statistics after taking the course
$x_1=$ dummy variable taking the value 1 if the 3-week course was taken and 0 if the 14 -week course was taken
$x_2=$ student's grade point average
$x_3=$ dummy variable taking the value 0 or 1 , depending on which of two teachers had taught the course
$x_4=$ dummy variable taking the value 1 if the student is male and 0 if female
$x_5=$ score on a standardized test of understanding of mathematics before taking the course
$x_6=$ number of semester credit hours the student had completed
$x_7=$ age of student
The numbers in parentheses under the coefficients are the estimated coefficient standard errors.

Write a report discussing what can be learned from this fitted regression.

Check back soon!
01:58

Problem 71

In a survey of 27 undergraduates at the University of Illinois the accompanying results were obtained with grade point averages $(y)$, the number of hours per week spent studying $\left(x_1\right)$, the average number of hours spent preparing for tests $\left(x_2\right)$, the number of hours per week spent in bars $\left(x_3\right)$, whether students take notes or mark highlights when reading texts ( $x_4=1$ if yes, 0 if no ), and the average number of credit hours taken per semester $\left(x_5\right)$. Estimate the regression of grade point average on the five independent variables, and write a report on your findings. The data are in the data file Student Performance.

Arwa Ali
Arwa Ali
Numerade Educator

Problem 72

You have been asked to develop a model to analyze salary in a large business organization.
The data for this model are stored in the file named Salorg; the variable names are self-explanatory.
a. Using the data in the file, develop a regression model that predicts salary as a function of the variables you select. Compute the conditional $F$ and conditional $t$ statistics for the coefficient of each predictor variable included in the model. Show all work and carefully explain your analysis process.
b. Test the hypothesis that female employees have a lower annual salary conditional on the variables in your model. The variable "Gender_1F" is coded 1 for female employees and 0 for male employees.
c. Test the hypothesis that the female employees have had a lower rate of salary increase conditional on the variables in the model developed for part $b$.

Check back soon!
01:31

Problem 73

Suppose that two independent variables are included as predictor variables in a multiple regression analysis. What can you expect will be the effect on the estimated slope coefficients when these two variables have each of the given correlations?
a. 0.78
b. 0.08
c. 0.94
d. 0.33

Jerrah Biggerstaff
Jerrah Biggerstaff
Numerade Educator
03:31

Problem 74

Consider a regression analysis with $n=34$ and four potential independent variables. Suppose that one of the independent variables has a correlation of 0.23 with the dependent variable. Does this imply that this independent variable will have a very small Student's $t$ statistic in the regression analysis with all four predictor variables?

Sneha Ravi
Sneha Ravi
Numerade Educator
03:31

Problem 75

Consider a regression analysis with $n=47$ and three potential independent variables. Suppose that one of the independent variables has a correlation of 0.95 with the dependent variable. Does this imply that this independent variable will have a very large Student's $t$ statistic in the regression analysis with all three predictor variables?

Sneha Ravi
Sneha Ravi
Numerade Educator
01:35

Problem 76

Consider a regression analysis with $n=49$ and two potential independent variables. Suppose that one of the independent variables has a correlation of 0.56 with the dependent variable. Does this imply that this independent variable will have a very small Student's $t$ statistic in the regression analysis with both predictor variables?

Nick Johnson
Nick Johnson
Numerade Educator
11:40

Problem 77

In order to assess the effect in one state of a casualty insurance company's economic power on its political power, the following model was hypothesized and fitted to data from all 50 states:
$$
Y=\beta_0+\beta_1 X_1+\beta_2 X_2+\beta_3 X_3+\beta_4 x_4+\beta_5 X_5+\varepsilon
$$
where
$$
\begin{aligned}
Y= & \text { ratio of company's payments for state and local } \\
& \text { taxes, in thousands of dollars, to total state and } \\
& \text { local tax revenues in millions of dollars } \\
X_1= & \text { insurance company state concentration ratio } \\
& \text { (a measure of the concentration of banking } \\
& \text { resources) }
\end{aligned}
$$
$$
\begin{aligned}
X_2= & \text { per capita income in the state in thousands } \\
& \text { of dollars } \\
X_3= & \text { ratio of nonfarm income to the sum of farm } \\
& \text { and nonfarm income } \\
X_4= & \text { ratio of insurance company's net after-tax in- } \\
& \text { come to insurance reserves (multiplied by } 1,000 \text { ) } \\
X_5= & \text { average of insurance reserves (divided by } \\
& 10,000)
\end{aligned}
$$
Part of the computer output from the estimated regression is shown here. Write a report summarizing the findings of this study.
$$
R \text {-Square }=0.515
$$
$$
\begin{array}{|c|c|c|c|}
\hline \text { Parameter } & \text { Estimate } & \begin{array}{c}
\text { Student's } t \\
\text { for } H_0: \\
\text { Parameter }=0
\end{array} & \begin{array}{c}
\text { Std. Error of } \\
\text { Estimate }
\end{array} \\
\hline \text { Intercept } & 10.60 & 2.41 & 4.40 \\
\hline \mathrm{X}_1 & -0.90 & -0.69 & 1.31 \\
\hline \mathrm{X}_2 & 0.14 & 0.50 & 0.28 \\
\hline X_3 & -12.85 & -2.83 & 4.18 \\
\hline X_4 & 0.080 & 0.50 & 0.160 \\
\hline \times 5 & 0.100 & 5.00 & 0.020 \\
\hline
\end{array}
$$

David Gagnon
David Gagnon
Numerade Educator
View

Problem 78

A random sample of 93 freshmen at the University of Illinois was asked to rate, on a scale of 1 (low) to 10 (high), their overall opinion of residence hall life. They were also asked to rate their levels of satisfaction with roommates, with the floor, with the hall, and with the resident advisor. (Information on satisfaction with the room itself was obtained, but this was later discarded as it provided no useful additional power in explaining overall opinion.) The following model was estimated:
$$
Y=\beta_0+\beta_1 X_1+\beta_2 X_2+\beta_3 X_3+\beta_4 X_4+\varepsilon
$$
where
$$
\begin{aligned}
Y & =\text { overall opinion of residence hall } \\
X_1 & =\text { satisfaction with roommates } \\
X_2 & =\text { satisfaction with floor } \\
X_3 & =\text { satisfaction with hall } \\
X_4 & =\text { satisfaction with resident advisor }
\end{aligned}
$$
Use the accompanying portion of the computer output from the estimated regression to write a report summarizing the findings of this study.
Dependent Variable: $Y$ Overall Opinion

$$
\begin{array}{lrcccc}
\hline & & \text { Sum of } & \text { Mean } & & \\
\text { Source } & \text { DF } & \text { Squares } & \text { Square } & F \text { Value } & R \text {-Square } \\
\hline \text { Model } & 4 & 37.016 & 9.2540 & 9.958 & 0.312 \\
\text { Error } & 88 & 81.780 & 0.9293 & & \\
\text { Total } & 92 & 118.79 & & & \\
\hline
\end{array}
$$
$$
\begin{array}{lccc}
\hline \text { Parameter } & \text { Estimate } & \begin{array}{c}
\text { Student's } t \\
\text { for } H_0:
\end{array} & \begin{array}{c}
\text { Std. Error of } \\
\text { Estimate }
\end{array} \\
\hline \text { Intercept } & 3.950 & 5.84 & 0.676 \\
\mathrm{X}_1 & 0.106 & 1.69 & 0.063 \\
\mathrm{X}_2 & 0.122 & 1.70 & 0.072 \\
\mathrm{X}_3 & 0.092 & 1.75 & 0.053 \\
\mathrm{X}_4 & 0.169 & 2.64 & 0.064 \\
\hline
\end{array}
$$

Shu Naito
Shu Naito
Numerade Educator
01:27

Problem 79

The following model was fitted to 47 monthly observations in an attempt to explain the difference between certificate of deposit rates and commercial paper rates:
$$
Y=\beta_0+\beta_1 X_1+\beta_2 X_2+\varepsilon
$$
where
$$
\begin{aligned}
Y= & \text { commercial paper certificate of deposit rate } \\
& \text { less commercial paper rate } \\
X_1= & \text { commercial paper rate } \\
X_2= & \text { ratio of loans and investments to capital }
\end{aligned}
$$
Use the part of the computer output from the estimated regression shown here to write a report summarizing the findings of this analysis.
$$
R \text {-Square }=0.730
$$

$$
\begin{array}{lccc}
\hline & & \begin{array}{c}
\text { Student's } \\
\text { for } H_0:
\end{array} & \begin{array}{c}
\text { Std. Error of } \\
\text { Estimate }
\end{array} \\
\hline \text { Parameter } & \text { Estimate } & \text { parameter }=0 & 1.343 \\
\mathrm{X}_1 & -5.559 & -4.14 & 0.033 \\
\mathrm{X}_2 & 0.186 & 5.64 & 0.216 \\
\hline
\end{array}
$$

Heather Duong
Heather Duong
Numerade Educator
05:14

Problem 80

You have been asked to develop a multiple regression model to predict the traffic fatality rate per 100 million miles in 2007. The data file Vehicle Travel State contains traffic data by state for the year 2007; the variables are described in the Chapter 11 appendix. Consider the following possible predictor variables and select only those that are conditionally significant; per capita disposable income, percent of population in urban areas, total licensed drivers, total motor vehicle registrations, percent interstate highway miles, motor vehicle fuel tax in cents per gallon, total highway expenditure divided by number of licensed drivers, doctors per 1,000 population, nurses per 1,000 population, and Medicaid enrollment as a fraction of total population.

Jon Southam
Jon Southam
Numerade Educator

Problem 81

The data file Economic Activity contains data for the 50 states in the United States; the variables are described in the Chapter 11 appendix. You are asked to develop a model to predict the percentage of females that are in the labor force. The possible predictor variables are per capita disposable personal income, the percentage of males unemployed, the manufacturing payroll per worker, and the unemployment rate of women $\left(x_3\right)$. Compute the multiple regression and write a report on your findings.

Check back soon!
02:21

Problem 82

The United Nations has hired you as a consultant to help identify factors that predict manufacturing growth in developing countries. You have decided to use multiple regression to develop a model and identify important variables that predict growth. You have collected the data in the data file Developing Country from 48 countries. The variables included are percentage manufacturing growth $(y)$, percentage agricultural growth $\left(x_1\right)$, percentage exports growth $\left(x_2\right)$, and percentage rate of inflation $\left(x_3\right)$ in 48 developing countries. Develop the multiple regression model and write a report on your findings.

James Kiss
James Kiss
Numerade Educator

Problem 83

The method of least squares is used far more often than any alternative procedure to estimate the parameters of a multiple regression model. Explain the basis for this method of estimation, and discuss why its use is so widespread.

Check back soon!
View

Problem 84

It is common practice to compute an analysis of variance table in conjunction with an estimated multiple regression. Carefully explain what can be learned from such a table.

Rashmi Sinha
Rashmi Sinha
Numerade Educator
00:48

Problem 85

State whether each of the following statements is true or false.
a. The error sum of squares must be smaller than the regression sum of squares.
b. Instead of carrying out a multiple regression, we can get the same information from simple linear regressions of the dependent variable on each independent variable.
c. The coefficient of determination cannot be negative.
d. The adjusted coefficient of determination cannot be negative.
e. The coefficient of multiple correlation is the square root of the coefficient of determination.

Erika Bustos
Erika Bustos
Numerade Educator
03:04

Problem 86

If an additional independent variable, however irrelevant, is added to a multiple regression model, a smaller sum-of-squared errors will result. Explain why this is so, and discuss the consequences for the interpretation of the coefficient of determination.

Prashant Bana
Prashant Bana
Numerade Educator
01:04

Problem 87

A dependent variable is regressed on two independent variables. It is possible that the hypotheses $H_0: \beta_1=0$ and $H_0: \beta_2=0$ cannot be rejected at low significance levels, yet the hypothesis $H_0: \beta_1=\beta_2=0$ can be rejected at a very low significance level. In what circumstances might this result arise?

Prashant Bana
Prashant Bana
Numerade Educator
View

Problem 88

[This exercise requires the material in the chapter appendix.] Suppose that the regression model
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\varepsilon
$$
is estimated by least squares. Show that the residuals, $e_{i,}$ from the fitted model sum to 0 .

Ana Carolina Da Cruz
Ana Carolina Da Cruz
Numerade Educator
View

Problem 89

A study was conducted to assess the influence of various factors on the start of new firms in the computer chip industry. For a sample of 70 countries the following model was estimated:
$$
\begin{aligned}
& \hat{y}=-59.31+4.983 x_1+\underset{(1.156)}{2.198 x_2}+\underset{(0.210)}{3.816 x_3}-\underset{(2.063)}{0.310 x_4} \\
& \text { (1.156) } \\
& \text { (0.210) } \\
& (2.063) \\
& -0.886 x_5+3.215 x_6+0.85 x_7 \\
& R^2=0.766 \\
&
\end{aligned}
$$
$(1.568)$
$(0.354)$
$$
R^2=0.766
$$
where
$\hat{y}=$ new business starts in the industry
$x_1=$ population in millions
$x_2=$ industry size
$x_3=$ measure of economic quality of life
$x_4=$ measure of political quality of life
$x_5=$ measure of environmental quality of life
$x_6=$ measure of health and educational quality of life
$x_7=$ measure of social quality of life
The numbers in parentheses under the coefficients are the estimated coefficient standard errors.
a. Interpret the estimated regression coefficients.
b. Interpret the coefficient of determination.
c. Find a $90 \%$ confidence interval for the increase in new business starts resulting from a one-unit increase in the economic quality of life, with all other variables unchanged.
d. Test, against a two-sided alternative at the $5 \%$ level, the null hypothesis that, all else remaining equal, the environmental quality of life does not influence new business starts.
e. Test, against a two-sided alternative at the $5 \%$ level, the null hypothesis that, all else remaining equal, the health and educational quality of life does not influence new business starts.
f. Test the null hypothesis that, taken together, these seven independent variables do not influence new business starts.

Shu Naito
Shu Naito
Numerade Educator
View

Problem 90

A survey research group conducts regular studies of households through mail questionnaires and is concerned about the factors influencing the response rate. In an experiment, 30 sets of questionnaires were mailed to potential respondents. The regression model fitted to the resulting data set was as follows:
$$
Y=\beta_0+\beta_1 X_1+\beta_2 X_2+\varepsilon
$$
where
$$
\begin{aligned}
Y & =\text { percentage of responses received } \\
X_1 & =\text { number of questions asked } \\
X_2 & =\text { length of questionnaire in number of words }
\end{aligned}
$$
Part of the SAS computer output from the estimate regression is shown next.
$$
R \text {-Square }=0.637
$$
$$
\begin{array}{|c|c|c|c|}
\hline \text { Parameter } & \text { Estimate } & \begin{array}{c}
\text { Student's } t \\
\text { for } H_0: \\
\text { Parameter }=0
\end{array} & \begin{array}{c}
\text { Std. Error of } \\
\text { Estimate }
\end{array} \\
\hline \text { Intercept } & 74.3652 & x^2= & 5 \\
\hline \mathrm{X} 1 & -1.8345 & -2.89 & 0.6349 \\
\hline \mathrm{X}_2 & -0.0162 & -1.78 & 0.0091 \\
\hline
\end{array}
$$
a. Interpret the estimated regression coefficients.
b. Interpret the coefficient of determination.
c. Test, at the $1 \%$ significance level, the null hypothesis that, taken together, the two independent variables do not linearly influence the response rate.
d. Find and interpret a $99 \%$ confidence interval for $\beta_1$.
e. Test the null hypothesis
$$
H_0: \beta_2=0
$$
against the alternative
$$
H_1: \beta_2<0
$$
and interpret your findings.

Victor Salazar
Victor Salazar
Numerade Educator

Problem 91

A consulting group offers courses in financial management for executives. At the end of these courses, participants are asked to provide overall ratings of the value of the course. To assess the impact of various factors on ratings, the model
$$
Y=\beta_0+\beta_1 X_1+\beta_2 X_2+\beta_3 X_3+\varepsilon
$$
was fitted for 25 such courses, where
$Y=$ average rating by participants of the course
$X_1=$ percentage of course time spent in group discussion sessions
$X_2=$ amount of money (in dollars) per course member spent on the preparation of subject matter material
$X_3=$ amount of money per course member spent on the provision of non-course-related material (food, drinks, and so for th)
Part of the SAS computer output for the fitted regression is shown next.

$$
\begin{array}{|c|c|c|c|}
\hline \text { Parameter } & \text { Estimate } & \begin{array}{c}
\text { Student's } t \\
\text { for } H_0 \text { : } \\
\text { Parameter }=0
\end{array} & \begin{array}{c}
\text { Std. Error of } \\
\text { Estimate }
\end{array} \\
\hline \text { Intercept } & 42.9712 & & \\
\hline \mathrm{X}_1 & 0.3817 & 1.89 & 0.2018 \\
\hline \mathrm{X}_2 & 0.5172 & 2.64 & 0.1957 \\
\hline \mathrm{X} 3 & 0.0753 & 1.09 & 0.0693 \\
\hline
\end{array}
$$

a. Interpret the estimated regression coefficients.
b. Interpret the coefficient of determination.
c. Test, at the $5 \%$ level, the null hypothesis that, taken together, the three independent variables do not linearly influence the course rating.
d. Find and interpret a $90 \%$ confidence interval for $\beta_1$.
e. Test the null hypothesis
$$
H_0: \beta_2=0
$$
against the alternative
$$
H_1: \beta_2>0
$$
and interpret your result.
f. Test at the $10 \%$ level the null hypothesis
$$
H_0: \beta_3=0
$$
against the alternative
$$
H_1: \beta_3 \neq 0
$$
and interpret your result.

Check back soon!
01:02

Problem 92

At the end of classes professors are rated by their students on a scale of 1 (poor) to 5 (excellent). Students are also asked what course grades they expect, and these are coded as $\mathrm{A}=4, \mathrm{~B}=3$, and so on. The data file Teacher Rating contains, for a random sample of 20 classes, ratings of professors, the average expected grades, and the numbers of students in the classes. The variables are defined in the data file. Compute the multiple regression of rating on expected grade and number of students, and write a report on your findings.

Sheryl Ezze
Sheryl Ezze
Numerade Educator

Problem 93

lyer Computer, Inc., wishes to know the effect of various variables on labor efficiency. Based on a sample of 64 observations, the following model was estimated by least squares:
$$
\begin{aligned}
\hat{y}= & -16.528+28.729 x_1+.022 x_2-0.023 x_3-0.054 x_4 \\
& -0.077 x_5+0.411 x_6+0.349 x_7+0.028 x_8 \quad R^2=.467
\end{aligned}
$$
where
$\hat{y}=$ index of direct labor efficiency in production plant
$x_1=$ ratio of overtime hours to straight-time hours worked by all production workers
$x_2=$ average number of hourly workers in the plant
$x_3=$ percentage of employees involved in some quality-of-work-life program
$x_4=$ number of grievances filed per 100 workers
$x_5=$ disciplinary action rate
$x_6=$ absenteeism rate for hourly workers
$x_7=$ salaried workers' attitudes, from low (dissatisfied) to high, as measured by questionnaire
$x_8=$ percentage of hourly employees submitting at least one suggestion in a year to the plant's suggestion program
Also obtained by least squares from these data was the fitted model:
$$
\hat{y}=9.062-10944 x_1+0.320 x_2+0.019 x_3 \quad R^2=0.242
$$
The variables $x_4, x_5, x_6, x_7$, and $x_8$ are measures of the performance of a plant's industrial relations system. Test, at the $1 \%$ level, the null hypothesis that they do not contribute to explaining direct labor efficiency, given that $x_1, x_2$ and $x_3$ are also to be used.

Check back soon!

Problem 94

Based on 107 students' scores on the first examination in a course on business statistics, the following model was estimated by least squares:
$$
\begin{aligned}
\hat{y} & =2.178+\underset{(0.090)}{0.469 x_1}+\underset{(0.456)}{3.369 x_2}+\underset{(1.457)}{3.054 x_3} \\
R^2 & =.686
\end{aligned}
$$
where
$\hat{y}=$ student's actual score on the examination
$x_1=$ student's expected score on the examination
$x_2=$ hours per week spent working on the course
$x_3=$ student's grade point average
The numbers in parentheses under the coefficients are the estimated coefficient standard errors.
a. Interpret the estimate of $\beta_1$.
b. Find and interpret a $95 \%$ confidence interval for $\beta_2$.
c. Test, against a two-sided alternative, the null hypothesis that $\beta_3$ is 0 , and interpret your result.
d. Interpret the coefficient of determination.
e. Test the null hypothesis that $\beta_1=\beta_2=\beta_3=0$.
f. Find and interpret the coefficient of multiple correlation.
g. Predict the score of a student who expects a score of 80 , works 8 hours per week on the course, and has a grade point average of 3.0.

Check back soon!
02:57

Problem 95

Based on 25 years of annual data, an attempt was made to explain savings in India. The model fitted was as follows:
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\varepsilon
$$
where
$$
\begin{aligned}
y & =\text { change in real deposit rate } \\
x_1 & =\text { change in real per capita income } \\
x_2 & =\text { change in real interest rate }
\end{aligned}
$$
The least squares parameter estimates (with standard errors in parentheses) were (Ghatak and Deadman $1989)$ as follows:
$$
b_1=0.0974(0.0215) \quad b_2=0.374(0.209)
$$
The adjusted coefficient of determination was as follows:
$$
\bar{R}^2=.91
$$
a. Find and interpret a $99 \%$ confidence interval for $\beta_1$.
b. Test, against the alternative that it is positive, the null hypothesis that $\beta_2$ is 0 .
c. Find the coefficient of determination.
d. Test the null hypothesis that $\beta_1=\beta_2=0$.
e. Find and interpret the coefficient of multiple

James Kiss
James Kiss
Numerade Educator
01:10

Problem 96

Based on data on 2,679 high school basketball players, the following model was fitted:
$$
y=\beta_0+\beta_1 x_1+\beta_2 x_2+\cdots+\beta_9 x_9+\varepsilon
$$
where
$$
\begin{aligned}
y & =\text { minutes played in season } \\
x_1 & =\text { field-goal percentage } \\
x_2 & =\text { free-throw percentage } \\
x_3 & =\text { rebounds per minute } \\
x_4 & =\text { points per minute } \\
x_5 & =\text { fouls per minute } \\
x_6 & =\text { steals per minute } \\
x_7 & =\text { blocked shots per minute } \\
x_8 & =\text { turnovers per minute } \\
x_9 & =\text { assists per minute }
\end{aligned}
$$

Erika Bustos
Erika Bustos
Numerade Educator
02:57

Problem 97

Based on data from 63 counties, the following model was estimated by least squares:
where
$\hat{y}=$ growth rate in real gross domestic product
$x_1=$ real income per capita
$x_2=$ average tax rate, as a proportion of gross national product
The numbers in parentheses under the coefficients are the estimated coefficient standard errors.
a. Test against a two-sided alternative the null hypothesis that $\beta_1$ is 0 . Interpret your result.
b. Test against a two-sided alternative the null hypothesis that $\beta_2$ is 0 . Interpret your result.
c. Interpret the coefficient of determination.
d. Find and interpret the coefficient of multiple correlation.

James Kiss
James Kiss
Numerade Educator
03:25

Problem 98

The following regression model was fitted to data on 60 U.S. female amateur golfers:
$$
\begin{aligned}
& \hat{y}=\underset{(10059)}{164,683}+\underset{(167.18)}{341.10 x_1}+\underset{(355.48)}{170.02 x_2}+\underset{(90.0)}{495.19 x_3}-\underset{\left(23 x_4\right.}{4.0} \\
& -136,040 x_5-35,549 x_6+202.52 x_7 \\
& (25.634) \\
& \overline{R^2}=.516 \\
&
\end{aligned}
$$
$(16,240)$
(106.20)
where
$\hat{y}=$ winnings per tournament in dollars
$x_1=$ average length of drive in yards
$x_2=$ percentage times drive ends in fairway
$x_3=$ percentage times green reached in regulation
$x_4=$ percentage times par saved after hitting into sand trap
$x_5=$ average number of putts taken on greens reached in regulation
$$
\begin{aligned}
x_6= & \text { average number of putts taken on greens not } \\
& \text { reached in regulation } \\
x_7= & \text { number of years the golfer has played }
\end{aligned}
$$
The numbers in parentheses under the coefficients are the estimated coefficient standard errors.

Write a report summarizing what can be learned from these results.

Shu Naito
Shu Naito
Numerade Educator

Problem 99

The Economics Department wishes to develop a multiple regression model to predict student GPA for economics courses. Department faculty have collected data for 112 graduates, which include the variables economics GPA, SAT verbal, SAT mathematics, ACT English, ACT social science, and high school percentile rank. The data are stored in a file named Student GPA on your data disk and described in the Chapter 11 appendix.
a. Use the SAT variables and class rank to determine the best prediction model. Remove any independent variables that are not significant. What are the coefficients, their Student's $t$ statistics, and the model?
b. Use the ACT variables and class rank to determine the best prediction model. Remove any independent variables that are not significant. What are the coefficients, their Student's $t$ statistics, and the model?
c. Which model predicts an economics GPA better? Present the evidence to support your conclusion.

Check back soon!
01:40

Problem 100

Use the data in the file Citydatr to estimate a regression equation that can be used to determine the marginal effect of the percent of commercial property on the market value per owner-occupied residence. Include the percent of owner-occupied residences, the percent of industrial property, the median number of rooms per residence, and the per capita income as additional predictor variables in your multiple regression equation. The variables are included on your data disk and described in the chapter appendix. Indicate which of the variables are conditionally significant. Your final equation should include only significant variables. Discuss and interpret your final regression model, including an indication of how you would select a community for your house.

Sheryl Ezze
Sheryl Ezze
Numerade Educator
05:07

Problem 101

The administrator of the National Highway Traffic Safety Administration (NHTSA) wants to know if the different types of vehicles in a state have a relationship to the highway death rate in the state. She has asked you to develop multiple regression analyses to determine if the average vehicle weight, the percentage of imported cars, the percentage of light trucks, and the average car age are related to crash deaths in automobiles and pickups. The data for the analysis are located in the data file named Vehicle Travel State. A description of the variables is contained in the Chapter 11 appendix.
a. Prepare a correlation matrix for crash deaths and the predictor variables. Note the simple relationships between crash deaths and the predictor
variables. In addition, indicate any potential multicollinearity problems between the predictor variables.
b. Prepare a multiple regression analysis of crash deaths on the potential predictor variables. Remove any nonsignificant predictor variables, one at a time, from the regression model. Indicate your best final model.
c. State the conclusions from your analysis and discuss the conditional importance of the variables in terms of their relationship to crash deaths.

James Kiss
James Kiss
Numerade Educator
01:24

Problem 102

The Department of Transportation wishes to know if states with a larger percentage of urban population have higher automobile and pickup crash death rates. In addition, it wants to know if the variable average speed on rural roads or the variable percentage of rural roads that are surfaced is conditionally related to crash death rates, given percentage of urban population. Data for this study are included in the file Vehicle Travel State; the variables are defined in the Chapter 11 appendix.
a. Prepare a correlation matrix and descriptive statistics for crash deaths and the potential predictor variables. Note the relationships and any potential problems of multicollinearity.
b. Prepare a multiple regression analysis of crash deaths on the potential predictor variables. Determine which of the variables should be retained in the regression model because they have a conditionally significant relationship.
c. State the results of your analysis in terms of your final regression model. Indicate which variables are conditionally significant.

Sheryl Ezze
Sheryl Ezze
Numerade Educator
02:20

Problem 103

An economist wishes to predict the market value of owner-occupied homes in small midwestern cities. He has collected a set of data from 45 small cities for a 2-year period and wants you to use this as the data source for the analysis. The data are in the file Citydatr the variables are described in the chapter appendix. He wants you to develop a multiple regression prediction equation. The potential predictor variables include the size of the house, tax rate, percent of commercial property, per capita income, and total city government expenditures.
a. Compute the correlation matrix and descriptive statistics for the market value of residences and the potential predictor variables. Note any potential problems of multicollinearity. Define the approximate range for your regression model by the variable means \pm 2 standard deviations.
b. Prepare multiple regression analyses using the predictor variables. Remove any variables that are not conditionally significant. Which variable, size of house or tax rate, has the stronger conditional relationship to the value of houses?
c. A business developer in a midwestern state has stated that local property tax rates in small towns need to be lowered because, if they are not, no one will purchase a house in these towns. Based on your analysis in this problem, evaluate the business developer's claim.

Dominador Tan
Dominador Tan
Numerade Educator
03:56

Problem 104

Stuart Wainwright, the vice president of purchasing for a large national retailer, has asked you to prepare an analysis of retail sales by state. He wants to know if the percent of unemployment for males and for females and the per capita disposable income are jointly related to the per capita retail sales. Data for this study are in the data file named Economic Activity; the variables are described in the Chapter 11 appendix. You may have to compute additional variables using the variables in the data file.
a. Prepare a correlation matrix, compute descriptive statistics, and obtain a regression analysis of per capita retail sales on unemployment and personal income. Compute $95 \%$ confidence intervals for the slope coefficients in each regression equation.
b. What is the conditional effect of a $\$ 1,000$ decrease in per capita income on per capita sales?
c. Would the prediction equation be improved by adding the state population as an additional predictor variable?

Sheryl Ezze
Sheryl Ezze
Numerade Educator

Problem 105

A major national supplier of building materials for residential construction is concerned about total sales for next year. It is well known that the company's sales are directly related to the total national residential investment. Several New York bankers are predicting that interest rates will rise about two percentage points next year. You have been asked to develop a regression analysis that can be used to predict the effect of interest rate changes on residential investment. In addition to interest rate, you believe that the GDP, money supply, government spending, and price index for finished goods might be predictors of residential investment. Therefore, you decide that two multiple regression models will be needed. One will include prime interest rate and important additional variables. The second will include federal funds interest rate and important additional variables. The time-series data for this study are contained in the data file named Macro2010, which is described in the Chapter 13 appendix.
a. Develop two multiple regression models to predict residential investment using prime interest rate for one and federal funds interest rate for the other. The final regression models should include only predictor variables that have a significant conditional effect. Analyze the regression statistics and indicate which equation provides the best predictions.
b. Determine the $95 \%$ confidence interval for the interest rate conditional slope coefficient in both regression equations.

Check back soon!
05:59

Problem 106

The Center for Disease Control (CDC) is interested in knowing if there are state-level population characteristics that predict the occurrence of breast cancer death rates and the occurrence of lung cancer death rates. The data file Staten, whose variables are described in the chapter appendix, contains a number of variables that could be possible predictors when used in combination. Your task is to develop multiple regression models that will determine which of the $K$ variables in the data file predict the breast cancer death rate and which predict the lung cancer death rate. Interpret your final regression model, including a discussion of the coefficients, their Student's $t^{\prime}$ s, the standard error of the estimate, and $R^2$.

Robin Corrigan
Robin Corrigan
Numerade Educator
03:04

Problem 107

You have been hired as a consultant to analyze the salary structure of Energy Futures, Inc., a firm that produces designs for solar energy applications. The company has operated for a number of years, and in recent years there have been an increasing number of complaints that the salaries paid to various workers. You have been provided data in the file Salary Study, whose variables are described in the Chapter 12 appendix. Your task is to determine the relationship between the various measures for each employee and the salary paid using a multiple regression analysis.

One particular complaint of great concern to the management is that female workers are paid less than male workers with the same experience and skill level. Test the hypothesis that the actual salary paid female workers and the rate of change in female salaries as a function of experience is less than the rate of change for male salaries as a function of experience. Your hypothesis test should be set up to provide strong evidence of discrimination against females if it exists. The test should be made conditional on the other significant predictor variables in your model.

Sheryl Ezze
Sheryl Ezze
Numerade Educator

Problem 108

Use the data in the data file named Student GPA, which is described in the Chapter 11 appendix, to develop a model to predict a student's grade point average in economics. Begin with the variables ACT scores, gender, and HSpct.
a. Use appropriate statistical procedures to choose a subset of statistically significant predictor variables.
Describe your strategy and carefully define your final model.
b. Discuss how this model might be used as part of the college's decision process to select students for admission.

Check back soon!
01:03

Problem 109

You have been asked to develop a model that will predict home prices as a function of important economic variables. After considerable research, you locate the work of Prof. Robert Shiller, Princeton University. Shiller has compiled data for housing costs beginning in 1890 . The data file Shiller House Price Cost is obtained from his data. The indexes for home price and building cost are developed to adjust for price changes over time. You are to develop a model using the Shiller data. Prepare a short interpretation of your model results. Variables are identified in the data file.
a. Does your model exhibit any tendency to predict high or low over the long time period? What is your evidence?
b. There was a housing price bubble in the first part of the 21st century. How could you identify this bubble using your model?

Carson Merrill
Carson Merrill
Numerade Educator

Problem 110

A major real estate developer has asked you to determine the effect of the interval between house sales, and the initial house sales price on second or final sales price with adjustments for the four major U.S. market areas identified in the data set. The data on housing prices are stored in the data file House Selling Price from the work of Robert Shiller. The data set includes the first and second sales price and the relative date of the house sales. Write a short report on the results of your analysis.

Check back soon!
01:03

Problem 111

A group of activists in Peaceful, Montana, are seeking increased development for this pristine enclave, which has received some national recognition on the television program Four Dirty Old Men. The group claims that increased commercial and industrial development will bring new prosperity and lower taxes to Peaceful. Specifically, it claims that an increased percentage of commercial and industrial development will decrease the property tax rate and increase the market value for owner-occupied residences.

You have been hired to analyze their claims. For this purpose you have obtained the data file Citydatr, which contains data from 45 small cities. The variables are described in the chapter appendix. From these data you will first develop regression models that predict the average value of owner-occupied housing and the property tax rate. Then you will determine if and how the addition of the percent of commercial property and then the percent of industrial property affects the variability in these regression models. The basic model for predicting market value of houses includes the size of house, the tax rate, the per capita income, and the percent of owner-occupied residences as independent variables. The basic model for predicting tax rate includes the tax assessment base, current city expenditures per capita, and the percent of owner-occupied residences as independent variables.
Determine if the percent of commercial and the percent of industrial variables improve the explained variability in each of the two models. Perform a conditional $F$ test for each of these additional variables. First, estimate the conditional effect of percent commercial property by itself and then the conditional effect of percent industrial property by itself. Carefully explain the results of your analysis. Include in your report an explanation of why it was important to include all the other variables in the regression model instead of just examining the effect of the direct and simple relationship between percent of commercial property and percent of industrial property on the tax rate and market value of housing.

Dominador Tan
Dominador Tan
Numerade Educator

Problem 112

You have been asked to develop a model that will predict the percentage of students who graduate in 4 years from highly ranked private colleges. The data file Private Colleges contains data collected by a national news service; descriptions of the predictor variables are contained in the Chapter 12 appendix.
a. Specify a list of potential predictor variables with a short rationale for each variable.
b. Use multiple regression to determine the conditional effect of each of these potential predictor variables.
c. Eliminate those variables that do not have a significant conditional effect to obtain your final model.
d. Prepare a short discussion regarding the conditional effects of the predictor variables in your model, based on your analysis.

Check back soon!

Problem 113

You have been asked to develop a model that will predict the cost with financial aid for students at highly ranked private colleges. The data file Private Colleges contains data collected by a national news service. Variables are identified in the Chapter 12 appendix.
a. Specify a list of potential predictor variables with a short rationale for each variable.
b. Use multiple regression to determine the conditional effect of each of these potential predictor variables.
c. Eliminate those variables that do not have a significant conditional effect to obtain your final model.
d. Prepare a short discussion regarding the conditional effects of the predictor variables in your model, based on your analysis.

Check back soon!

Problem 114

You are asked to develop a multiple regression model that indicates the relationship between a person's physical characteristics and the quality of diet consumed as measured by the Healthy Eating Index (HEI-2005). The predictor variables to be used are a doctor's diagnosis of high blood pressure (doc bp), the ratio of waist measure to obese waist measure (waistper), the body mass index (BMI), whether the subject was overweight (sr overweight), male compared to female (female), and age (age). Also, the model should include a dummy variable to indicate the effect of first versus the second interview.
a. Estimate the model using the basic specification variables indicated here.
b. Estimate the model again, but in this case include a variable that adjusts for immigrant versus native person (immigrant).
c. Estimate the model again, but in this case include a variable that adjusts for single status versus a person with a partner (single).
d. Estimate the model again, but in this case include a variable that adjusts for participation in the food stamp program (fsp).

Check back soon!

Problem 115

You are asked to develop a multiple regression model that indicates the relationship between a person's behavioral characteristics and the quality of diet consumed as measured by the Healthy Eating Index (HEI-2005). The predictor variables to be used are whether subject limited weight (sr did lm wt), whether the subject was a smoker (smoker), number of hours subject spent in front of a TV or computer screen (screen hours), sedentary versus active subject (activity level; note you will need to recode to a dummy variable), percent of subject's calories from a fast-food restaurant (pff), percent of subject's calories eaten at home ( $\mathrm{P}$ ate at Home), whether subject was a college graduate (col grad), and subject's household income (hh income est). Also, the model should include a dummy variable to indicate the effect of first versus second interview.
a. Estimate the model using the basic specification variables indicated here.
b. Estimate the model again. but in this case include a variable that adjusts for immigrant versus native person (immigrant).
c. Estimate the model again, but in this case include a variable that adjusts for single status versus a person with a partner (single).
d. Estimate the model again, but in this case include a variable that adjusts for participation in the food stamp program (fsp).

Check back soon!

Problem 116

You are asked to develop a multiple regression model that indicates the relationship between a person's physical characteristics and the daily cost of food (daily cost). The predictor variables to be used are a doctor's diagnosis of high blood pressure (doc bp), the ratio of waist measure to obese waist measure (waistper), the body mass index (BMI), whether the subject was overweight (sr overweight), male compared to female (female), and age (age). Also, the model should include a dummy variable to indicate the effect of first versus the second interview.
a. Estimate the model using the basic specification variables indicated here.
b. Estimate the model again, but in this case include a variable that adjusts for immigrant versus native person (immigrant).
c. Estimate the model again, but in this case include a variable that adjusts for single status versus a person with a partner (single).
d. Estimate the model again, but in this case include a variable that adjusts for participation in the food stamp program (fsp).

Check back soon!

Problem 117

You are asked to develop a multiple regression model that indicates the relationship between a person's behavioral characteristics and the daily cost of food (daily cost). The predictor variables to be used are subject's limiting weight (sr did lm wt), subject being a smoker (smoker), subject's number of hours in front of a TV or computer screen (screen hours), subject's being sedentary versus active (activity level: note that you will need to recode to a dummy variable), percent of subject's calories from a fast-food restaurant (pff), percent of subject's calories eaten at home ( $\mathrm{P}$ ate at Home), whether the subject is a college graduate (col grad), and household income (hh income est). Also, the model should include a dummy variable to indicate the effect of first versus second interview.
a. Estimate the model using the basic specification variables indicated here.
b. Estimate the model again, but in this case include a variable that adjusts for immigrant versus native person (immigrant).
c. Estimate the model again but in this case include a variable that adjusts for single status versus a person with a partner (single).
d. Estimate the model again, but in this case include a variable that adjusts for participation in the food stamp program (fsp).

Check back soon!