Using a Dummy

4 Key excerpts on "Using a Dummy"

• 

  eBook - PDF

  Handbook of Data Analysis

  • Melissa A Hardy, Alan Bryman, Melissa A Hardy, Alan Bryman(Authors)
  • 2009(Publication Date)
  • SAGE Publications Ltd

    (Publisher)

  The number of dummy vari-ables used to represent the classification scheme is determined by how many cat-egories the researcher wants to keep distinct. If the original classification contains too much detail, researchers can combine cat-egories, so long as they are careful to ensure that these combinations are theoretically and empirically similar. Later chapters address techniques for analyzing qualitative dependent variables in single-equation and multiple-equation models. In this chapter, we will concentrate on ways to maximize information by using dummy variables as exogenous variables in standard regression analysis. We will assume that the reader is familiar with the coding techniques discussed in Chapter 2. Our purpose here is to describe issues of statistical analysis and interpretation when dummy variables are specified. In the first section of the chapter, we will move from the simplest specification of a model with one binary classification vari-able to models with one multinomial classifi-cation variable, then multiple classification variables, ending with a model that includes both dummy variables and interval variables. 9 Incorporating Categorical Information into Regression Models: The Utility of Dummy Variables MELISSA HARDY AND JOHN REYNOLDS Once we have elaborated the basic interpretations of these partial effects, we will expand the specification to include inter-action terms so that we can test whether rela-tionships are uniform across subgroups. We compare the interaction models to separate subgroup regression models and demonstrate the similarities and differences between the two approaches. We conclude this section of the chapter with an illustration of the decom-position of group differences.

  Sign up to read

  Learn more about book
• 

  eBook - PDF

  Introductory Econometrics

  A Practical Approach

  • Hamid Seddighi(Author)
  • 2013(Publication Date)
  • Routledge

    (Publisher)

  In recent years, due to the availability of large-scale micro-econometric data, these models have become popular, particularly in the area of market research. This introductory chapter deals with the qualitative variables as regressors in regression models, while the next chapter provides an introductory discussion of the qualitative depen- dent variable models. Key topics • Dummy variables • Seasonal adjustment of data using dummy variables • Pooling time series and cross-section data • Testing for the structural break using dummy variables 9.1 Dummy variables and regression analysis Dummy variables are used in regression analysis to take account of the impact of qualitative variables on the dependent variable. They are binary variables taking only the values of unity 228 Qualitative variables in econometric models – panel data regression models and zero. Unity is used to denote the occurrence of an event/characteristic, while zero signi- fies the absence of a qualitative characteristic. Dummy variables are used in cross-section and time series regression models. In a time series model they are frequently employed for seasonal adjustments of data. In a cross- section analysis they are normally employed to capture the impact of a qualitative variable, such as gender or level of education, on the dependent variable. We now demonstrate the application of the dummy variables in regression/econometric models, using a number of examples. 9.1.1. A model of household consumption expenditure, cross section data: measuring the impact of location on household consumption expenditure The location of households can have a significant impact on household consumption. Location is a qualitative variable and to capture its impacts, we need to employ dummy variables. In the example of family consumption expenditure, to bring the location variable into the analysis, we distinguish between two locations: south and north.

  Sign up to read

  Learn more about book
• 

  eBook - ePub

  Applied Regression Analysis

  • Norman R. Draper, Harry Smith(Authors)
  • 2014(Publication Date)
  • Wiley-Interscience

    (Publisher)

  CHAPTER 14 “Dummy” Variables

  What Are “Dummy” Variables?

  The variables considered in regression equations usually can take values over some continuous range. Occasionally we must introduce a factor that has two or more distinct levels. For example, data may arise from three machines, or two factories, or six operators. In such a case we cannot set up a continuous scale for the variable “machine” or “factory” or “operator.” We must assign to these variables some levels in order to take account of the fact that the various machines or factories or operators may have separate deterministic effects on the response. Variables of this sort are usually called dummy variables. They are usually (but not always) unrelated to any physical levels that might exist in the factors themselves.

  One example of a dummy variable is found in the attachment of a variable X 0 (whose value is always unity) to the term β 0 in a regression model. The X 0 is unnecessary but provides a notational convenience at times. Other dummy variables are somewhat more than a mere convenience, as we shall see.

  An Infinite Number of Choices

  The suggestions we make for setting up dummy variable systems are not unique. Typically, there are an infinite number of alternative ways to set up a system to cover any particular type of situation. Given a particular selection of dummy variable vectors that “works,” that is, represents the factors as desired, we can derive other sets by taking linear combinations of the vectors in the first set. As long as the second set is chosen so that their vectors are not linearly dependent on one another, all will be well. In general, the most useful dummy variable setups are simple in form, employing levels of 0 and 1, for example, or -1 and 1. Usefulness, however, lies in the eye of the user (as “beauty lies in the eye of the beholder”).

  14.1. DUMMY VARIABLES TO SEPARATE BLOCKS OF DATA WITH DIFFERENT INTERCEPTS, SAME MODEL

  Sign up to read

  Learn more about book
• 

  eBook - PDF

  Applied Regression Analysis

  • Norman R. Draper, Harry Smith(Authors)
  • 2014(Publication Date)
  • Wiley-Interscience

    (Publisher)

  C HAP T E R 14 "Dummy" Variables What Are "Dummy" Variables? The variables considered in regression equations usually can take values over some continuous range. Occasionally we must introduce a factor that has two or more distinct levels. For example, data may arise from three machines, or two factories, or six operators. In such a case we cannot set up a continuous scale for the variable "machine" or "factory" or "operator." We must assign to these variables some levels in order to take account of the fact that the various machines or factories or operators may have separate deterministic effects on the response. Variables of this sort are usually called dummy variables. They are usually (but not always) unrelated to any physical levels that might exist in the factors themselves. One example of a dummy variable is found in the attachment of a variable Xo (whose value is always unity) to the term f30 in a regression model. The Xu is unnecessary but provides a notational convenience at times. Other dummy variables are somewhat more than a mere convenience, as we shall see. An Infinite Number of Choices The suggestions we make for setting up dummy variable systems are not unique. Typically, there are an infinite number of alternative ways to set up a system to cover any particular type of situation. Given a particular selection of dummy variable vectors that "works," that is, represents the factors as desired, we can derive other sets by taking linear combinations of the vectors in the first set. As long as the second set is chosen so that their vectors are not linearly dependent on one another, all will be well. In general, the most useful dummy variable setups are simple in form, employing levels of 0 and 1, for example, or ~ I and 1. Usefulness, however, lies in the eye of the user (as "beauty Iies in the eye of the beholder").

  Sign up to read

  Learn more about book