EMET2007 Lecture 7

# They involve the squared residuals from the

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

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: icity, but heteroscedasticity robust versions are available in most software Also called White/Eicker standard errors. They involve the squared residuals from the regression and from a regression of xj on all other explanatory variables. Lecture 7 (heteroscedasticity) EMET2007/6007 24 th April 2013 32 / 34 Example: Hourly wage equation \ log (wage ) = 0.128 + 0.0904educ + 0.0410exper (0.105 ) [0.107 ] b = wage ε (0.0075 ) [0.0078 ] (0.0052 ) [0.0050 ] 0.007 exper 2 (0.0001 ) [0.0001 ] wage the residuals [ Heteroscedasticity robust standard errors (in square brackest ) may be larger or smaller than their nonrobust counterparts. The di¤erences are often small in practice. H0 F Lecture 7 (heteroscedasticity) : βex per = βex per 2 = 0 = 17.95 and Frobust = 17.99 EMET2007/6007 24 th April 2013 33 / 34 In this lecture we have covered ... Heteroscedasticity De…nition Consequences Detecting by anecdotal evidence and formal testing Improving estimation under heteroscedasticity (WLS and FGLS) Adjusting the standard errors to make them robust to heteroscedasticity Lecture 7 (heteroscedasticity) EMET2007/6007 24 th April 2013 34 / 34...
View Full Document

## This note was uploaded on 06/15/2013 for the course EMET 2007 taught by Professor Strachan during the Two '13 term at Australian National University.

Ask a homework question - tutors are online