Exam 12: Simple Linear Regression

SCENARIO 12-13 In this era of tough economic conditions, voters increasingly ask the question: "Is the educational achievement level of students dependent on the amount of money the state in which they reside spends on education?" The partial computer output below is the result of using spending per student ($) as the independent variable and composite score which is the sum of the math, science and reading scores as the dependent variable on 35 states that participated in a study.The table includes only partial results. $\begin{array}{l}\begin{array} { l c } \hline { \text { Regression Statistics } } \\\hline \text { Multiple R } & 0.3122 \\\text { R Square } & 0.0975 \\\text { Adjusted R } & 0.0701 \\\text { Square } & \\\text { Standard } & 26.9122 \\\text { Error } & \\\text { Observations } & 35 \\\hline\end{array}\\\text { ANOVA }\\\begin{array}{lrrr}&df&SS&MS&F\\\hline\text { Regression } & 1 & 2581.5759 & \\\text { Residual } & & & 724.2674 \\\text { Total } & 34 & 26482.4000 &\\\hline\end{array}\\\\\begin{array} { l c c c c } \hline & \text { Coefficients } & \text { Standard Error } & t \text { Stat } & P \text {-value } \\\hline \text { Intercept } & 595.540251 & 22.115176 & & \\\text { Spending per } & & & & \\\text { Student(\$) } & 0.007996 & 0.004235 & \\\hline\end{array}\end{array}$ -Referring to Scenario 12-13, the p-value of the measured F-test statistic to test whether spending per student affects composite score is .

If the correlation coefficient (r) = 1.00, then

SCENARIO 12-10 The management of a chain electronic store would like to develop a model for predicting the weekly sales (in thousands of dollars) for individual stores based on the number of customers who made purchases.A random sample of 12 stores yields the following results: Customers Sales (Thousands of Dollars) 907 11.20 926 11.05 713 8.21 741 9.21 780 9.42 898 10.08 510 6.73 529 7.02 460 6.12 872 9.52 650 7.53 603 7.25 -Referring to Scenario 12-10, construct a 95% prediction interval for the weekly sales of a store that has 600 purchasing customers.

SCENARIO 12-5 The managing partner of an advertising agency believes that his company's sales are related to the industry sales.He uses Microsoft Excel to analyze the last 4 years of quarterly data with the following results: Regression Statistics Multiple R 0.802 R Square 0.643 Adjusted R Square 0.618 Standard Error SYx 0.9224 Observations 16 ANOVA df SS MS F Sig.F Regression 1 21.497 21.497 25.27 0.000 Error 14 11.912 0.851 Total 15 33.409 Predictor Coef StdError tStat P-value Intercept 3.962 1.440 2.75 0.016 Industry 0.040451 0.008048 5.03 0.000 $\text { Durbin-Watson Statistic } \quad 1.59$ -Referring to Scenario 12-5, the partner wants to test for autocorrelation using the Durbin-Watson statistic.Using a level of significance of 0.05, the critical values of the test are dL = , and dU = _.

SCENARIO 12-4 The managers of a brokerage firm are interested in finding out if the number of new clients a broker brings into the firm affects the sales generated by the broker.They sample 12 brokers and determine the number of new clients they have enrolled in the last year and their sales amounts in thousands of dollars.These data are presented in the table that follows. Broker Clients Sales 1 27 52 2 11 37 3 42 64 4 33 55 5 15 29 6 15 34 7 25 58 8 36 59 9 28 44 10 30 48 11 17 31 12 22 38 -Referring to Scenario 12-4, the regression sum of squares (SSR) is .

SCENARIO 12-13 In this era of tough economic conditions, voters increasingly ask the question: "Is the educational achievement level of students dependent on the amount of money the state in which they reside spends on education?" The partial computer output below is the result of using spending per student ($) as the independent variable and composite score which is the sum of the math, science and reading scores as the dependent variable on 35 states that participated in a study.The table includes only partial results. $\begin{array}{l}\begin{array} { l c } \hline { \text { Regression Statistics } } \\\hline \text { Multiple R } & 0.3122 \\\text { R Square } & 0.0975 \\\text { Adjusted R } & 0.0701 \\\text { Square } & \\\text { Standard } & 26.9122 \\\text { Error } & \\\text { Observations } & 35 \\\hline\end{array}\\\text { ANOVA }\\\begin{array}{lrrr}&df&SS&MS&F\\\hline\text { Regression } & 1 & 2581.5759 & \\\text { Residual } & & & 724.2674 \\\text { Total } & 34 & 26482.4000 &\\\hline\end{array}\\\\\begin{array} { l c c c c } \hline & \text { Coefficients } & \text { Standard Error } & t \text { Stat } & P \text {-value } \\\hline \text { Intercept } & 595.540251 & 22.115176 & & \\\text { Spending per } & & & & \\\text { Student(\$) } & 0.007996 & 0.004235 & \\\hline\end{array}\end{array}$ -Referring to Scenario 12-13, the conclusion on the test of whether spending per student affects composite score using a 5% level of significance is

SCENARIO 12-3 The director of cooperative education at a state college wants to examine the effect of cooperative education job experience on marketability in the work place.She takes a random sample of 4 students.For these 4, she finds out how many times each had a cooperative education job and how many job offers they received upon graduation.These data are presented in the table below. Student Coop Jobs Job Offer 1 1 4 2 2 6 3 1 3 4 0 1 -Referring to Scenario 12-3, the total sum of squares (SST) is .

SCENARIO 12-11 A computer software developer would like to use the number of downloads (in thousands) for the trial version of his new shareware to predict the amount of revenue (in thousands of dollars) he can make on the full version of the new shareware.Following is the output from a simple linear regression along with the residual plot and normal probability plot obtained from a data set of 30 different sharewares that he has developed: Regression Statistics Multiple R 0.8691 R Square 0.7554 Adjusted R Square 0.7467 Standard Error 44.4765 Observations 30.0000 ANOVA df SS MS F Significance F Regression 1 171062.9193 171062.9193 86.4759 0.0000 Residual 28 55386.4309 1978.1582 Total 29 226451.3503 Coefficients Standard Error t Stat P-value Lower 95\% Upper 95\% Intercept -95.0614 26.9183 -3.5315 0.0015 -150.2009 -39.9218 Download 3.7297 0.4011 9.2992 0.0000 2.9082 4.5513 Simple Linear Regression 12-41 -Referring to Scenario 12-11, the null hypothesis for testing whether there is a linear relationship between revenue and the number of downloads is "There is no linear relationship between revenue and the number of downloads".

SCENARIO 12-1 A large national bank charges local companies for using their services.A bank official reported the results of a regression analysis designed to predict the bank's charges (Y) -- measured in dollars per month -- for services rendered to local companies.One independent variable used to predict service charges to a company is the company's sales revenue (X) -- measured in millions of dollars.Data for 21 companies who use the bank's services were used to fit the model: $Y _ { i } = \beta _ { 0 } + \beta _ { 1 } X _ { i } + \varepsilon _ { i }$ Theresultsofthesimplelinearregressionareprovidedbelow. $\bar { I } = - 2,700 + 20 X , S _ { Z X } = 65 \text {, two-tail } p \text { value } = 0.034 \text { (for testing } \beta _ { 1 } \text { ) }$ -Referring to Scenario 12-1, interpret the estimate of $\beta$ , the standard deviation of the random error term (standard error of the estimate) in the model.

SCENARIO 12-10 The management of a chain electronic store would like to develop a model for predicting the weekly sales (in thousands of dollars) for individual stores based on the number of customers who made purchases.A random sample of 12 stores yields the following results: Customers Sales (Thousands of Dollars) 907 11.20 926 11.05 713 8.21 741 9.21 780 9.42 898 10.08 510 6.73 529 7.02 460 6.12 872 9.52 650 7.53 603 7.25 -Referring to Scenario 12-10, what is the p-value of the t test statistic when testing whether the number of customers who make a purchase affects weekly sales?

SCENARIO 12-13 In this era of tough economic conditions, voters increasingly ask the question: "Is the educational achievement level of students dependent on the amount of money the state in which they reside spends on education?" The partial computer output below is the result of using spending per student ($) as the independent variable and composite score which is the sum of the math, science and reading scores as the dependent variable on 35 states that participated in a study.The table includes only partial results. $\begin{array}{l}\begin{array} { l c } \hline { \text { Regression Statistics } } \\\hline \text { Multiple R } & 0.3122 \\\text { R Square } & 0.0975 \\\text { Adjusted R } & 0.0701 \\\text { Square } & \\\text { Standard } & 26.9122 \\\text { Error } & \\\text { Observations } & 35 \\\hline\end{array}\\\text { ANOVA }\\\begin{array}{lrrr}&df&SS&MS&F\\\hline\text { Regression } & 1 & 2581.5759 & \\\text { Residual } & & & 724.2674 \\\text { Total } & 34 & 26482.4000 &\\\hline\end{array}\\\\\begin{array} { l c c c c } \hline & \text { Coefficients } & \text { Standard Error } & t \text { Stat } & P \text {-value } \\\hline \text { Intercept } & 595.540251 & 22.115176 & & \\\text { Spending per } & & & & \\\text { Student(\$) } & 0.007996 & 0.004235 & \\\hline\end{array}\end{array}$ -Referring to Scenario 12-13, the decision on the test of whether composite score depends linearly on spending per student using a 10% level of significance is to (reject or not reject) H0 .

SCENARIO 12-11 A computer software developer would like to use the number of downloads (in thousands) for the trial version of his new shareware to predict the amount of revenue (in thousands of dollars) he can make on the full version of the new shareware.Following is the output from a simple linear regression along with the residual plot and normal probability plot obtained from a data set of 30 different sharewares that he has developed: Regression Statistics Multiple R 0.8691 R Square 0.7554 Adjusted R Square 0.7467 Standard Error 44.4765 Observations 30.0000 ANOVA df SS MS F Significance F Regression 1 171062.9193 171062.9193 86.4759 0.0000 Residual 28 55386.4309 1978.1582 Total 29 226451.3503 Coefficients Standard Error t Stat P-value Lower 95\% Upper 95\% Intercept -95.0614 26.9183 -3.5315 0.0015 -150.2009 -39.9218 Download 3.7297 0.4011 9.2992 0.0000 2.9082 4.5513 Simple Linear Regression 12-41 -Referring to Scenario 12-11, predict the revenue when the number of downloads is 30 thousand.

SCENARIO 12-9 It is believed that, the average numbers of hours spent studying per day (HOURS) during undergraduate education should have a positive linear relationship with the starting salary (SALARY, measured in thousands of dollars per month) after graduation.Given below is the Excel output for predicting starting salary (Y) using number of hours spent studying per day (X) for a sample of 51 students.NOTE: Only partial output is shown. Regression Statistics Multiple R 0.8857 R Square 0.7845 Adjusted R Square 0.7801 Standard Error 1.3704 Observations 51 ANOVA df SS MS F Significance F Regression 1 335.0472 335.0473 178.3859 Residual 1.8782 Total 50 427.0798 Coefficients Standard Error t Stat P -value Lower 95\% Upper 95\% Intercept -1.8940 0.4018 -4.7134 0.0000 -2.7015 -1.0865 Hours 0.9795 0.0733 13.3561 0.0000 0.8321 1.1269 Note: 2.051E - 05 = 2.051 *10-05 and 5.944 E - 18 =5.944 *10-18 . -Referring to Scenario 12-9, the estimated change in mean salary (in thousands of dollars)as a result of spending an extra hour per day studying is

SCENARIO 12-13 In this era of tough economic conditions, voters increasingly ask the question: "Is the educational achievement level of students dependent on the amount of money the state in which they reside spends on education?" The partial computer output below is the result of using spending per student ($) as the independent variable and composite score which is the sum of the math, science and reading scores as the dependent variable on 35 states that participated in a study.The table includes only partial results. $\begin{array}{l}\begin{array} { l c } \hline { \text { Regression Statistics } } \\\hline \text { Multiple R } & 0.3122 \\\text { R Square } & 0.0975 \\\text { Adjusted R } & 0.0701 \\\text { Square } & \\\text { Standard } & 26.9122 \\\text { Error } & \\\text { Observations } & 35 \\\hline\end{array}\\\text { ANOVA }\\\begin{array}{lrrr}&df&SS&MS&F\\\hline\text { Regression } & 1 & 2581.5759 & \\\text { Residual } & & & 724.2674 \\\text { Total } & 34 & 26482.4000 &\\\hline\end{array}\\\\\begin{array} { l c c c c } \hline & \text { Coefficients } & \text { Standard Error } & t \text { Stat } & P \text {-value } \\\hline \text { Intercept } & 595.540251 & 22.115176 & & \\\text { Spending per } & & & & \\\text { Student(\$) } & 0.007996 & 0.004235 & \\\hline\end{array}\end{array}$ -Referring to Scenario 12-13, the p-value of the measured t-test statistic to test whether composite score depends linearly on spending per student is .

SCENARIO 12-13 In this era of tough economic conditions, voters increasingly ask the question: "Is the educational achievement level of students dependent on the amount of money the state in which they reside spends on education?" The partial computer output below is the result of using spending per student ($) as the independent variable and composite score which is the sum of the math, science and reading scores as the dependent variable on 35 states that participated in a study.The table includes only partial results. $\begin{array}{l}\begin{array} { l c } \hline { \text { Regression Statistics } } \\\hline \text { Multiple R } & 0.3122 \\\text { R Square } & 0.0975 \\\text { Adjusted R } & 0.0701 \\\text { Square } & \\\text { Standard } & 26.9122 \\\text { Error } & \\\text { Observations } & 35 \\\hline\end{array}\\\text { ANOVA }\\\begin{array}{lrrr}&df&SS&MS&F\\\hline\text { Regression } & 1 & 2581.5759 & \\\text { Residual } & & & 724.2674 \\\text { Total } & 34 & 26482.4000 &\\\hline\end{array}\\\\\begin{array} { l c c c c } \hline & \text { Coefficients } & \text { Standard Error } & t \text { Stat } & P \text {-value } \\\hline \text { Intercept } & 595.540251 & 22.115176 & & \\\text { Spending per } & & & & \\\text { Student(\$) } & 0.007996 & 0.004235 & \\\hline\end{array}\end{array}$ -Referring to Scenario 12-13, the value of the F test on whether spending per student affects composite score is .

The slope (b1) represents

SCENARIO 12-4 The managers of a brokerage firm are interested in finding out if the number of new clients a broker brings into the firm affects the sales generated by the broker.They sample 12 brokers and determine the number of new clients they have enrolled in the last year and their sales amounts in thousands of dollars.These data are presented in the table that follows. Broker Clients Sales 1 27 52 2 11 37 3 42 64 4 33 55 5 15 29 6 15 34 7 25 58 8 36 59 9 28 44 10 30 48 11 17 31 12 22 38 -Referring to Scenario 12-4, the coefficient of correlation is .

SCENARIO 12-6 The following Excel tables are obtained when "Score received on an exam (measured in percentage points)" (Y) is regressed on "percentage attendance" (X) for 22 students in a Statistics for Business and Economics course. Regression Statistics Multiple R 0.142620229 R Square 0.02034053 Standard Error 20.25979924 Observations 22 Coefficients Standard Error T Stat P-value Intercept 39.39027309 37.24347659 1.057642216 0.302826622 Attendance 0.340583573 0.52852452 0.644404489 0.526635689 -Referring to Scenario 12-6, which of the following statements is true?

SCENARIO 12-4 The managers of a brokerage firm are interested in finding out if the number of new clients a broker brings into the firm affects the sales generated by the broker.They sample 12 brokers and determine the number of new clients they have enrolled in the last year and their sales amounts in thousands of dollars.These data are presented in the table that follows. Broker Clients Sales 1 27 52 2 11 37 3 42 64 4 33 55 5 15 29 6 15 34 7 25 58 8 36 59 9 28 44 10 30 48 11 17 31 12 22 38 -Referring to Scenario 12-4, the managers of the brokerage firm wanted to test the hypothesis that the population slope was equal to 0.For a test with a level of significance of 0.01, the null hypothesis should be rejected if the value of the test statistic is .

SCENARIO 12-5 The managing partner of an advertising agency believes that his company's sales are related to the industry sales.He uses Microsoft Excel to analyze the last 4 years of quarterly data with the following results: Regression Statistics Multiple R 0.802 R Square 0.643 Adjusted R Square 0.618 Standard Error SYx 0.9224 Observations 16 ANOVA df SS MS F Sig.F Regression 1 21.497 21.497 25.27 0.000 Error 14 11.912 0.851 Total 15 33.409 Predictor Coef StdError tStat P-value Intercept 3.962 1.440 2.75 0.016 Industry 0.040451 0.008048 5.03 0.000 $\text { Durbin-Watson Statistic } \quad 1.59$ -Referring to Scenario 12-5, the estimates of the Y-intercept and slope are and_, respectively.