WILMORE PAPER COMPANY

Advertising Budget Model

Time Series Analysis by Burcu Arikan, Spring 2008

Introduction Specification Estimation Verification Validation Budget

 

Specification

Between Fall 2000 and Winter 2008, sales of cases of paper towels at Wilmore Paper Company showed steady growth. The first step in time series analysis involves identification of the historical relationship between advertising expenditures and sales revenue. The resulting scotter plot is shown below.

 

The scatterplot gives us an estimate of the curve and since it clearly follows a non-linear regression model, we move on to the next step in our model building process.

In the analysis, nine response functions (see below) in the Current Effect model were analyzed with WPC's sales as dependent variable and WPC's advertising expenditures as independent variable. The accuracy of a function in predicting the dependent variable was tested by comparing the function's predicted values with the actual values from WPC's historical data.