Multiple Regression A – Tutorial Problem
Question 1
Fanfare International, Inc., designs, distributes, and markets ceiling fans and lighting fixtures. The
company’s product line includes 120 basic models of ceiling fans and 138 compatible fan light kits
and table lamps. These products are marketed to over 1000 lighting showrooms and electrical
wholesales that supply the remodeling and new construction markets. The product line is distributed
by a sales organization of 58 independent sales representatives.
In the summer of 2013, Fanfare decided it needed to develop forecasts of future sales to help
determine future salesforce needs, capital expenditures, and so on. The file named “T2AQ1.xlsx”
contains the year and month the data being collected, monthly sales data, as well as three additional
variables for the period July 2009 through April 2013. The variables are defined as follows
SALES = total monthly sales in thousands of dollars
ADEX = advertising expense in thousands of dollars
MTGRATE = mortgage rate for 30-year loans (%)
HSTARTS = housing starts in thousands of units
As a consultant to Fanfare, your job is to find a regression model to forecast future sales for Fanfare.
Decide which variables you should include in the equation and which should be omitted with
justification. How well do you believe the developed model will do at forecasting future sales?
Question 2
A random sample of 42 firms was chosen from the S&P 500 firms listed in Spring 2003 Special
Issue of Business Week. The indicated dividend yield (DIVYIELD), the earnings per share (EPS),
and the stock price (PRICE) were recorded for these 42 firms. These data are available in the file
named “T2AQ2.xlsx”. Run a regression using DIVYIELD as the response and EPS and PRICE as
the predictors. Use the output to answer the following questions.
a. What is the sample regression equation relating DIVYIELD to PRICE and EPS?
b. What percentage of the variation of DIVYIELD has been explained by the regression model?
c. Test the overall fit of the regression model. Use 10% level of significance.
d. Is it necessary to test each slope coefficient individually to see if either PRICE or EPS is related
to DIVYIEDL? Why or why not?
Question 3
The ABX Company is interested in conducting a study to identify factors that affect absenteeism
among its production employees. After reviewing the literatures on absenteeism and interviewing
several production supervisors and a number of employees, the researcher in charge of the project
defined the variables shown as below
ABSENT = the number of distinct occasions that the worker was absent during the year
COMPLX = a measure of job complexity, ranging from 0 to 100
PAY = basic hourly pay rate in dollars
SENIOR = number of complete years with the company
AGE = age of the employee
DEPEND = employee response to the question “how many individuals other than yourself
depend on you for most of their financial support?
A sample of 77 employees was randomly selected, the data was contained in the file named
“T2AQ3.xlsx”. The response is absenteeism, while other variables are considered as possible
explanatory variables.
Identify the factors that may relate to absenteeism at a 5% level of significance. Write down your
final model and justify your choice of variables in the model. Check to see if your choice of variables
and the estimated coefficients make intuitive sense. How much variation in absenteeism has been
explained? What does this tell you? Does your model give you some sense of which employees
might be absent most often? If so, which ones? What might be done to reduce absenteeism?
