Translating Strategy into Shareholder Value: A Company-Wide Approach to Value Creation

An Overview of Regression Analysis

AT ITS MOST BASIC LEVEL, regression analysis is a statistical technique for developing an equation that describes the relationship between two or more variables.^[1] This equation generally takes the form of:

The following variables are represented in the equation:

• Y equals the dependent variable.

• x_n equals the independent variables; the parameters used to predict y.

• a equals the value of y when all independent variables are zero.

• b₁ equals the regression coefficient that indicates the impact that the independent variables have on the dependent variable.

One variable, the dependent variable, is the variable that you are trying to understand. The dependent variable is the answer. The other variable(s), the independent or explanatory variables, are the factors you feel affect the (dependent) variable or the answer. Let's say a national general merchandise retailer may think that there is a direct relationship between sales and advertising expenditure; more specifically, that an increase in advertising expenditure will result in an increase in total sales. This is based on the rationale that increased advertising will produce greater brand awareness, leading to larger demand for the company's product. In this example, sales are dependent (what you are solving for)—at least to some degree—upon advertising expenditures.

^[1]Terry E. Dielman, Applied Regression Analysis for Business and Economics (PWS-Kent, 1991).