Question

Data on a sample of 100 customers of Alibaba, a direct marketing company, for the current year. The variables are defined as follows: Customer = a code for the customer Own Home = 1 if customer owns home, o if renting Close = 1 if lives close to stores (with similar merchandise), O if not Salary = annual household salary in $ Catalogs = number of catalogs this customer was sent this year Amount Spent = Amount spent total on purchases this year You are given the result of the multiple regression with dependent variable = Amount Spent. Use this output to answer the questions

(a) - (g) below the output. SUMMARY OUTPUT Regression Statistics Multiple R 0.8366 R Square 0.6999 Adjusted R Squ 0.6873 Standard Error 512.72 Observations 100 ANOVA df f ignificance F 55.39 0 .000 Regression Residual Total 4 95 99 SS 58241400.88 24974011.87 83215412.75 M S 14560350.22 262884.34 Intercept Own Home Close Salary Catalogs Coefficients Standard Error -91.67 185.917 -229.47 121.800 -604.90 114.229 0.02216 0.002 42.62 8.451 t Stat -0.493 -1.884 -5.295 9.836 5.043 P-value Lower 95% Upper 95% 0.623 -460.76 277.42 0.063 -471.27 12.34 0.000 -831.67 -378.12 0.000 0.02 0.03 0.000 25.84 59.40
(b) Using the regression model in (a), predict the mean amount spent by Amy, a customer who has the following characteristics: does not own a home, but lives close to stores that have similar merchandise, salary is $60,000 and is sent 12 catalogs this year. Give the value rounded to the nearest cent.
(c) Suppose Brenda has the same characteristics as Amy but does not live close to store with similar merchandise. How much would you expect Brenda to spend? Give the value rounded to the nearest cent.

Answer 1

a.
`\hat{amount} \ \hat{ spent} = -91.67 - (229.47* ownhome)-(604.90 * close) +(0.02216 * salary) + (42.62 * catalogs)`

b. the expected amount spent by Amy is $1144.47

c. the expected amount that Brenda is going to spend is $1749.37

Explanation:

(a)

From the regression output; the equation for the regression model can be written as:

`\hat{amount} \ \hat{ spent} = -91.67 - (229.47* ownhome)-(604.90 * close) +(0.02216 * salary) + (42.62 * catalogs)`

From the information given in the question;

(b)

Amy does not own a home but rent; the variables given also stated that ;

Own Home = 1 if customer owns home, 0 if renting

So for Amy ; Own Home = 0 (since it is rented)

Close = Yes(1)

Salary = $60,000

Catalogs = 12

Therefore;

the mean amount spent by Amy is by using the regression model is ;

`\hat{amount} \ \hat{ spent} = -91.67 - (229.47* 0)-(604.90 * 1) +(0.02216 * 60000) + (42.62 * 12)`

`\hat{amount} \ \hat{ spent} = -91.67 -0-604.90 +1329.6 + 511.44`

`\hat{amount} \ \hat{ spent} =1144.47`

Thus; the expected amount spent by Amy is $1144.47

(c)

If Brenda has the same characteristics as Amy but does not live close to store with similar merchandise.

Then the Close for Brenda will be = No (0)

Thus; the amount spent by Brenda will be:

`\hat{amount} \ \hat{ spent} = -91.67 - (229.47* 0)-(604.90 * 0) +(0.02216 * 60000) + (42.62 * 12)`

`\hat{amount} \ \hat{ spent} = -91.67 -0-0 +1329.6 + 511.44`

`\hat{amount} \ \hat{ spent} = 1749.37`

Thus, the expected amount that Brenda is going to spend is $1749.37