Using Dummy Variables in CER Development

Dummy variables (also referred to as indicator variables) are commonly used in regression analysis to stratify data into distinct categories. However, many analysts specify dummy variables in their cost estimating relationships (CER) without properly analyzing the statistical validity of using them. For example, the dummy variable t-test should be applied to determine the relevance of using dummy variables, but this test is often neglected. Consequently, the fit statistics can be misleading. The dummy variable t-test is useful for determining whether the slope (or exponent) coefficients in different categories are significantly different. This is directly applicable to the dummy variable CER where we assume distinct categories in the data set share the same sensitivity for the ordinary independent variable, while the only difference is in the response levels. This paper explains when to use dummy variables and how to use them correctly when deriving CERs. Specific guidelines are proposed to help analysts determine if the application of dummy variables is appropriate for their data set. This paper also demonstrates some common errors when applying dummy variables to real examples and explains how to use the Chow test and dummy variable t-test to validate the CER. An application using dummy variables in splines (to derive the fitted equation as well as the intersection) is also discussed. OUTLINE The main objectives of this paper are threefold. First, we discuss the purpose of using dummy variables and their properties in a regression equation. We will then describe using the Chow test and t-test for checking the significance of the overall model and the individual dummy variables, respectively. General concerns/pitfalls of using dummy variables are also discussed. Finally, we describe a common application of using dummy variables in Spline. The following topics will be discussed: