EMET2007 Lecture 9 for Wattle

4 percentage points lecture 9 qualitative variables

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 1, the expected house price increases by 5.4 percentage points ) Lecture 9 (qualitative variables) %∂price ∂ log (price ) = t 5.4% ∂colonial ∂colonial EMET2007/6007 8 st May 2013 18 / 53 Using dummy variables for multiple categories 1) De…ne membership in each category by a dummy variable 2) Leave out one category (which becomes the base category) For example, consider the categories: men and women, married and single. We have four groups: single women, married women, single men, married men We will set single men as the base category and include dummies for the other three Lecture 9 (qualitative variables) EMET2007/6007 8 st May 2013 19 / 53 \ log (wage ) = 3.21 (0.100 ) + 0.213 marrmale 0.198 marrfem (0.055 ) (0.0.058 ) + 0.079 educ + 0.027 exper (0.007 ) (0.005 ) + 0.029 tenure (0.007 ) n = 526 0.110 singfem (0.0.056 ) 2 0.00054exper (0.00011 ) 2 0.00053tenure (0.00023 ) 2 R = 0.364 Holding other things …xed, married women are expected to earn 19.8% less than single men (= the base category) Lecture 9 (qualitative variables) EMET2007/6007 8 st May 2013 20 / 53 Incorporating ordinal information using dummy variables Example: City credit ratings and municipal bond interest rates MBR = β0 + β1 CR + other factors MBR = Municipal bond rate CR = Credit rating from 0-4 (0=worst, 4=best so there are …ve categories) This speci…cation would probably not be appropriate as the credit rating only contains ordinal information. Lecture 9 (qualitative variables) EMET2007/6007 8 st May 2013 21 / 53 A better way to incorporate this information is to de…ne dummies: MBR = β0 + δ1 CR1 + δ2 CR2 + δ3 CR3 + δ4 CR4 + other factors Dummies indicating whether the particular rating applies, e.g. CR1=1 if CR=1 and CR1=0 otherwise. All e¤ects are measured in comparison to the worst rating (= base category). Lecture 9 (qualitative variables) EMET2007/6007 8 st May 2013 22 / 53 Interactions involving dummy variables Allowing for di¤erent slopes log (wage ) = β0 + δ0 female + β1 educ + δ1 female educ + ε β0 = intercept for men β1 = slope for men β0 + δ0 = intercept for women β1 + δ1 = slope for women Interesting hypotheses: The return to education is the same for men and women H0 : δ 1 = 0 The whole wage equation is the same for men and women H0 : δ0 = 0 and δ1 = 0 Lecture 9 (qualitative variables) EMET2007/6007 8 st May 2013 23 / 53 Interacting both the intercept and the slope with the female dummy enables one to model completely in...
View Full Document

Ask a homework question - tutors are online