This preview shows page 1. Sign up to view the full content.
Unformatted text preview: elated
to other factors a¤ecting the outcome. Lecture 9 (qualitative variables) EMET2007/6007 8 st May 2013 50 / 53 Further example of an endogenous dummy regressor
Are nonwhite customers discriminated against?
The dependent variable is:
approved = Dummy indicating whether loan was approved (1 =
approved, 0 = not approved) The explanatory variables include:
nonwhite = 1 Race dummy, = 0 if white
credrate = Credit rating It is important to control for other characteristics that may be
important for loan approval (e.g. profession, unemployment)
Omitting important characteristics that are correlated with the
non-white dummy will produce spurious evidence for discriminiation
Lecture 9 (qualitative variables) EMET2007/6007 8 st May 2013 51 / 53 I will be in Austria and Romania next week so there will be no lecture
Tutorials will proceed as usual: i.e., there will be tutorials next week,
and the week after, and the week after ... Lecture 9 (qualitative variables) EMET2007/6007 8 st May 2013 52 / 53 In this lecture we have covered ...
Qualitative, categorical and ordinal independent and dependent
The magic of the dummy variable
The dummy variable trap and not including a constant
Dummy variables giving intercept shifts
Dummy variables and log dependent variables
Dummies and Ordinal variables
Dummy variables giving slope shifts (interacting dummies and normal
Binary dependent variables and the linear probability model
Lecture 9 (qualitative variables) EMET2007/6007 8 st May 2013 53 / 53...
View Full Document
- Two '13