Multiple Regression - Using Two or More Independent Variables

In Simple Regression Analysis, the model consists of one dependent variable, and one independent variable.

Multiple Regression Analysis models use one dependent variable, but use two or more independent variables that the researcher believes have explanatory power to predict the value of the dependent variable.

Using your selected Business Research Project or a company/industry of your interest:

• Define a key performance indicator (for example, annual unit sales) that would serve as the dependent variable that you would like to predict. Give this variable a name that is eight characters or less in length, as you would use it in a statistical software package.

• Define two or more independent variables (also known as explanatory variables) that you believe would have predictive power to use in a multiple regression model to predict the value of the dependent variable. Give these variables names that are eight characters or less in length, as you would use it in a statistical software package.

• For each independent variable, (a) define what type of variable it is (quantitative or qualitative) and (b) how it would be measured. Remember, any qualitative variable should be dichotomous (meaning the attribute is either present or it is not) and you should indicate the anchor descriptions for the 0 or 1 values).

• Describe any potential challenges you think could be present in the design of your model.

• Please comment on the models presented by other learners to identify strengths and potential challenges of their designs.

What are some ways we can determine if adding a specific independent variable is improving the predictive power of our multiple regression model? How do we know if the independent variables we are using provide sufficient explanatory power?