This preview shows page 1. Sign up to view the full content.
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 Goodnessof…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 GaussMarkov assumptions
Onesided and twosided hypothesis tests on β1
Used tstatistics 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.
 Two '13
 strachan

Click to edit the document details