EMET2007 Lecture 4

0155salesi bi 00089 r 2 00144 89893 2629 log

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: the e¤ect is signi…cantly di¤erent from zero because zero is outside the interval. Lecture 4 (econometrics and simple linear regression) EMET2007/6007 13 th March 2013 47 / 50 Further on R 2 We can compare the R 2 for following two regressions \ salary i \ salary i = 1174 + 0.0155salesi + bi ε (0.0089 ) = R 2 = 0.0144 898.93 + 262.9 log (salesi ) + bi ε (92.36 ) R 2 = 0.0377 But we cannot compare the R 2 for the following two regressions \ salary i = 1174 + 0.0155salesi + bi ε (0.0089 ) c log (salaryi ) = 6.85 + 0.000015salesi + bi ε (3.55e 06 ) R 2 = 0.0144 R 2 = 0.0791 Why not? Lecture 4 (econometrics and simple linear regression) EMET2007/6007 13 th March 2013 48 / 50 Ok, I’ tell you ... ll To be able to compare the Goodness-of-…t of two models using R 2 , they must have the same dependent variable. If one regression has salary for the dependent variable and another has log (salary ) for the dependent variable, we cannot compare R 2 Lecture 4 (econometrics and simple linear regression) EMET2007/6007 13 th March 2013 49 / 50 In this lecture we have covered ... Introduced the assumption of normal errors This implied the estimators were normal Listed the Gauss-Markov assumptions One-sided and two-sided hypothesis tests on β1 Used t-statistics and p values Introduced the concept of statistical signi…cance Con…dence intervals and their interpretations Learned a little more about the use of R 2 Lecture 4 (econometrics and simple linear regression) EMET2007/6007 13 th March 2013 50 / 50...
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