Christopher Dougherty EC220 - Introduction to econometrics (chapter 2) Slideshow: exercise 2.16 Original citation: Dougherty, C.

Download Report

Transcript Christopher Dougherty EC220 - Introduction to econometrics (chapter 2) Slideshow: exercise 2.16 Original citation: Dougherty, C.

Christopher Dougherty
EC220 - Introduction to econometrics
(chapter 2)
Slideshow: exercise 2.16
Original citation:
Dougherty, C. (2012) EC220 - Introduction to econometrics (chapter 2). [Teaching Resource]
© 2012 The Author
This version available at: http://learningresources.lse.ac.uk/128/
Available in LSE Learning Resources Online: May 2012
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 License. This license allows
the user to remix, tweak, and build upon the work even for commercial purposes, as long as the user
credits the author and licenses their new creations under the identical terms.
http://creativecommons.org/licenses/by-sa/3.0/
http://learningresources.lse.ac.uk/
EXERCISE 2.16
2.16
A researcher with a sample of 50 individuals with similar
education but differing amounts of training hypothesizes
that hourly earnings, EARNINGS, may be related to hours
of training, TRAINING, according to the relationship
EARNINGS = b1 + b2 TRAINING + u
He is prepared to test the null hypothesis H0: b2 = 0 against
the alternative hypothesis H1: b2  0 at the 5 percent and 1
percent levels. What should he report
1.
2.
3.
4.
If b2 = 0.30, s.e.(b2) = 0.12?
If b2 = 0.55, s.e.(b2) = 0.12?
If b2 = 0.10, s.e.(b2) = 0.12?
If b2 = -0.27, s.e.(b2) = 0.12?
1
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
There are 50 observations and 2 parameters have been estimated, so there are 48 degrees
of freedom.
2
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
tcrit, 5% = 2.01, tcrit, 1% = 2.68
The table giving the critical values of t does not give the values for 48 degrees of freedom.
We will use the values for 50 as a guide. For the 5% level the value is 2.01, and for the 1%
level it is 2.68. The critical values for 48 will be slightly higher.
3
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
tcrit, 5% = 2.01, tcrit, 1% = 2.68
_______________________________________________
1. If b2 = 0.30, s.e.(b2) = 0.12?
t = 2.50.
In the first case, the t statistic is 2.50.
4
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
tcrit, 5% = 2.01, tcrit, 1% = 2.68
_______________________________________________
1. If b2 = 0.30, s.e.(b2) = 0.12?
t = 2.50. Reject H0 at the 5% level but not at the 1%
level.
This is greater than the critical value of t at the 5% level, but less than the critical value at
the 1% level.
5
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
tcrit, 5% = 2.01, tcrit, 1% = 2.68
_______________________________________________
1. If b2 = 0.30, s.e.(b2) = 0.12?
t = 2.50. Reject H0 at the 5%, but not at the 1%, level.
In this case we should mention both tests. It is not enough to say "Reject at the 5% level",
because it leaves open the possibility that we might be able to reject at the 1% level.
6
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
tcrit, 5% = 2.01, tcrit, 1% = 2.68
_______________________________________________
1. If b2 = 0.30, s.e.(b2) = 0.12?
t = 2.50. Reject H0 at the 5%, but not at the 1%, level.
Likewise it is not enough to say "Do not reject at the 1% level", because this does not reveal
whether the result is significant at the 5% level or not.
7
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
tcrit, 5% = 2.01, tcrit, 1% = 2.68
_______________________________________________
2. If b2 = 0.55, s.e.(b2) = 0.12?
t = 4.58.
In the second case, t is equal to 4.58.
8
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
tcrit, 5% = 2.01, tcrit, 1% = 2.68
_______________________________________________
2. If b2 = 0.55, s.e.(b2) = 0.12?
t = 4.58. Reject H0 at the 1% level.
We report only the result of the 1% test. There is no need to mention the 5% test. If you do,
you reveal that you do not understand that rejection at the 1% level automatically means
rejection at the 5% level, and you look ignorant.
9
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
tcrit, 5% = 2.01, tcrit, 1% = 2.68
_______________________________________________
2. If b2 = 0.55, s.e.(b2) = 0.12?
t = 4.58. Reject H0 at the 0.1% level (tcrit, 0.1% = 3.50).
Actually, given the large t statistic, it is a good idea to investigate whether we can reject H0
at the 0.1% level. It turns out that we can. The critical value for 50 degrees of freedom is
3.50. So we just report the outcome of this test. There is no need to mention the 1% test.
10
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
tcrit, 5% = 2.01, tcrit, 1% = 2.68
_______________________________________________
2. If b2 = 0.55, s.e.(b2) = 0.12?
t = 4.58. Reject H0 at the 0.1% level (tcrit, 0.1% = 3.50).
Why is it a good idea to press on to a 0.1% test, if the t statistic is large? Try to answer this
question before looking at the next slide.
11
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
tcrit, 5% = 2.01, tcrit, 1% = 2.68
_______________________________________________
2. If b2 = 0.55, s.e.(b2) = 0.12?
t = 4.58. Reject H0 at the 0.1% level (tcrit, 0.1% = 3.50).
The reason is that rejection at the 1% level still leaves open the possibility of a 1% risk of
having made a Type I error (rejecting the null hypothesis when it is in fact true). So there is
a 1% risk of the "significant" result having occurred as a matter of chance.
12
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
tcrit, 5% = 2.01, tcrit, 1% = 2.68
_______________________________________________
2. If b2 = 0.55, s.e.(b2) = 0.12?
t = 4.58. Reject H0 at the 0.1% level (tcrit, 0.1% = 3.50).
If you can reject at the 0.1% level, you reduce that risk to one tenth of 1%. This means that
the result is almost certainly genuine.
13
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
tcrit, 5% = 2.01, tcrit, 1% = 2.68
_______________________________________________
3. If b2 = 0.10, s.e.(b2) = 0.12?
t = 0.83.
In the third case, t is equal to 0.83.
14
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
tcrit, 5% = 2.01, tcrit, 1% = 2.68
_______________________________________________
3. If b2 = 0.10, s.e.(b2) = 0.12?
t = 0.83. Do not reject H0 at the 5% level.
We report only the result of the 5% test. There is no need to mention the 1% test. If you do,
you reveal that you do not understand that not rejecting at the 5% level automatically means
not rejecting at the 1% level, and you look ignorant.
15
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
tcrit, 5% = 2.01, tcrit, 1% = 2.68
_______________________________________________
4. If b2 = -0.27, s.e.(b2) = 0.12?
t = -2.25.
In the fourth case, t is equal to -2.25.
16
EXERCISE 2.16
EARNINGS = b1 + b2 TRAINING + u
H0: b2 = 0, H1: b2  0
n = 50, so 48 degrees of freedom
tcrit, 5% = 2.01, tcrit, 1% = 2.68
_______________________________________________
4. If b2 = -0.27, s.e.(b2) = 0.12?
t = -2.25. Reject H0 at the 5% level but not at the 1%
level.
The absolute value of the t statistic is between the critical values for the 5% and 1% tests.
So we mention both tests, as in the first case.
17
Copyright Christopher Dougherty 1999–2006. This slideshow may be freely copied for
personal use.
20.06.06