Definition:Dummy Variable

From ProofWiki
Jump to navigation Jump to search


A dummy variable is a quantity written in a mathematical expression in the form of a variable, even though it does in fact represent a constant.

It is usual for a dummy variable to take on the value either $0$ or $1$.


Regression Equation

Consider the regression equation with the two explanatory variables $x_1$ and $x_2$:

$y = \beta_0 + \beta_1 x_1 + \beta_2 x_2$

It is sometimes more convenient to write it as:

$y = \beta_0 x_0 + \beta_1 x_1 + \beta_2 x_2$

where in this context $x_0$ is a dummy variable which always takes the value $1$.

Hence, if we also want to consider the regression model where $\beta_0 = 0$, this can be accommodated by setting the dummy variable $x_0$ equal to $0$.

Also defined as

It is often the case that the term dummy variable is used to apply to what on $\mathsf{Pr} \infty \mathsf{fWiki}$ is called a bound variable.

This usage, while common, is deprecated on $\mathsf{Pr} \infty \mathsf{fWiki}$ to reduce the chance of ambiguity.

Also see

  • Results about dummy variables can be found here.