Unformatted text preview: ternatives (H0 : β1 0, H1 : β1 < 0) Reject the null hypothesis in favour of the alternative hypothesis if
the estimated coe¢ cient is “too small” (i.e. smaller  more negative than a critical value).
Again construct the critical value so that, if the null hypothesis is
true, it is rejected in, for example, 5% of the cases.
In the following example, this is the point of the tdistribution with
678 degrees of freedom that is exceeded in 5% of the cases. ! Reject if t statistic less than Lecture 4 (econometrics and simple linear regression) 1.645 EMET2007/6007 13 th March 2013 27 / 50 In this example we wish to know if the number of classes missed in a
semester (missed ) explains GPA scores (termGPA):
termGPAi
termGPAi = β0 + β1 missedi + εi
= b0 + b1 missedi + bi
β
β
ε
= 3.04 0.0043missedi + bi
ε The Stata output with n = 680, has R 2 = 0.3133 and Lecture 4 (econometrics and simple linear regression) EMET2007/6007 13 th March 2013 28 / 50 H0 : β 1 0 H1 : β 1 < 0 One would either expect a positive e¤ect of class attendance on GPA
tb =
β
1 Pr (t678 < b
β1 se b1
β = df = 680 0.0756
=
0.0043 17.59 2 = 678 1.645) = 0.05 and Pr (t678 < The null hypothesis is rejected since 17.59 < 2.33) = 0.01 1.645 Conclude that there is evidence that missing classes has a negative
e¤ect on expected GPA Lecture 4 (econometrics and simple linear regression) EMET2007/6007 13 th March 2013 29 / 50 Again with p values
H0 : β 1 0 H1 : β 1 < 0 One would either expect a positive e¤ect of class attendance on GPA
p 0.000
= 0.000 and tb = 17.59 < 0
β1
2
1.645) = 0.05 and Pr (t678 < 2.33) = 0.01 value = Pr (t678 < The null hypothesis is rejected since 0.000 < 0.05
Conclude that there is no evidence that return on earnings has any
e¤ect upon the CEO’ salary
s Lecture 4 (econometrics and simple linear regression) EMET2007/6007 13 th March 2013 30 / 50 Testing against twosided alternatives (H0 : β1 = 0, H1 : β1 6= 0)
Reject the null hypothesis in favour of the alternative hypothesis if
the absolute value of the estimated coe¢ cient is “too big” (i.e.
bigger than a critical value).
Again construct the critical value so that, if the null hypothesi...
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