The Constant Term
Functional Forms
One Tailed Tests for Unitary Elasticity
Chapter 7 Specification: Choosing a Functional Form Lecture 1
Masao Ogaki
Department of Economics, Ohio State University
November 6, 2007
Masao Ogaki Chapter 7 Specificat
Econ 444, Monday November 20, class 16
Econ 444, Monday November 20, class 16
Robert de Jong1
1 Department
of Economics Ohio State University
Robert de Jong
Econ 444, Monday November 20, class 16
Econ 444, Monday November 20, class 16
Monday No
Econ 444, Wednesday September 27, class 3
Econ 444, Wednesday September 27, class 3
Robert de Jong1
1 Department
of Economics Ohio State University
Robert de Jong
Econ 444, Wednesday September 27, class 3
Econ 444, Wednesday September 27, class
Econ 444, Monday September 25, class 2
Econ 444, Monday September 25, class 2
Robert de Jong1
1 Department
of Economics Ohio State University
Robert de Jong
Econ 444, Monday September 25, class 2
Econ 444, Monday September 25, class 2
First:
1
Econ 444, Wednesday September 20, class 1
Econ 444, Wednesday September 20, class 1
Robert de Jong1
1 Department
of Economics Ohio State University
Robert de Jong
Econ 444, Wednesday September 20, class 1
Econ 444, Wednesday September 20, class
Econ 444 Elementary Econometrics (Fall 2010)
HOMEWORK EXERCISE: 1
Due in class on Thursday, Jan 21 or Friday, Jan 22 Depending on When We Meet for the Make-up Class Q1. Choose the one alternative that best completes the statement or answers the question.
Econ 444 Elementary Econometrics (Winter 2010)
HOMEWORK EXERCISE: 2
Due in class on Wednesday, Jan 27 1. State with reasons whether the following statements are true or false. (1.a) The stochastic error term i and the residual ei mean the same thing, and
Econ 444 Elementary Econometrics (Winter 2010)
ANSWERS TO HOMEWORK EXERCISE: 2
1. (1.a) False. The stochastic error term i is for the population regression and is not directly observable unless the population regression is, while the residual is for the s
Econ 444 Elementary Econometrics (Winter 2010)
HOMEWORK EXERCISE: 3
Due in my Mail Box at Arps Hall Room 410, by 4:30 pm on Friday Feb 5. 1. MULTIPLE CHOICE. Choose the one alternative that best completes the statement or answers the question. 1)Heteroske
Econ 444, Wednesday October 4, class 4
Econ 444, Wednesday October 4, class 4
Robert de Jong1
1 Department
of Economics Ohio State University
Robert de Jong
Econ 444, Wednesday October 4, class 4
Econ 444, Wednesday October 4, class 4
1 2 3 4 5
Econ 444, Wednesday October 11, class 5
Econ 444, Wednesday October 11, class 5
Robert de Jong1
1 Department
of Economics Ohio State University
Robert de Jong
Econ 444, Wednesday October 11, class 5
Econ 444, Wednesday October 11, class 5
Class
Econ 444, Wednesday October 16, class 6
Econ 444, Monday October 16, class 6
Robert de Jong1
1 Department
of Economics Ohio State University
Robert de Jong
Econ 444, Monday October 16, class 6
Econ 444, Wednesday October 16, class 6
Class progr
Econ 444, Wednesday November 15, class 15
Econ 444, Wednesday November 15, class 15
Robert de Jong1
1 Department
of Economics Ohio State University
Robert de Jong
Econ 444, Wednesday November 15, class 15
Econ 444, Wednesday November 15, class 1
Econ 444, Monday November 13, class 14
Econ 444, Monday November 13, class 14
Robert de Jong1
1 Department
of Economics Ohio State University
Robert de Jong
Econ 444, Monday November 13, class 14
Econ 444, Monday November 13, class 14
Class pro
Class 12, Wednesday November 8
Econ 444, Wednesday November 8, class 12
Robert de Jong1
1 Department
of Economics Ohio State University
Wednesday November 8
Robert de Jong
Econ 444, Wednesday November 8, class 12
Class 12, Wednesday November 8
Econ 444, Wednesday November 1, class 10
Econ 444, class 11
Robert de Jong1
1 Department
of Economics Ohio State University
Monday November 6
Robert de Jong
Econ 444, class 11
Econ 444, Wednesday November 1, class 10
Monday November 6
1 2
Exe
Econ 444, Wednesday November 1, class 10
Robert de Jong1
1 Department
of Economics Ohio State University
Robert de Jong
Econ 444, Wednesday November 1, class 10
Monday November 1
1 2
Exercise for today: model assumptions New material
Robert de
Class 9, Monday October 30
Econ 444, class 9
Robert de Jong1
1 Department
of Economics Ohio State University
October 30, 2006
Robert de Jong
Econ 444, class 9
Class 9, Monday October 30
Example
DJ i = 0
+
0.10 (0.01)
Ri1
+
0.0010 (0.00
Econ 444, Econ 444, Wednesday October 25
Robert de Jong1
1 Department
of Economics Ohio State University
Robert de Jong
Econ 444, Econ 444, Wednesday October 25
Previous class: GRE scores: are GRE and SAT biased against women and ethnic groups?
G
Econ 444, Wednesday October 18, class 7
Econ 444, Wednesday October 18, class 7
Robert de Jong1
1 Department
of Economics Ohio State University
Robert de Jong
Econ 444, Wednesday October 18, class 7
Econ 444, Wednesday October 18, class 7
Wedne
Econ 444 Elementary Econometrics (Winter 2010)
ANSWERS TO HOMEWORK EXERCISE: 3
1. MULTIPLE CHOICE 1)C 2)A 3)D
2. a. See Eviews output below.
1
b. The sample regression function is : Yi = 4.855176 + 0.916609X1i + 0.158596X2i SE = (14.54034) (0.149560) (0.0
Econ 444 Elementary Econometrics (Winter 2010)
HOMEWORK EXERCISE: 4
Due in class on Wednesday, Feb 24. 1. Suppose we want to estimate, log (Yi ) = 1 +2 log (X2i )+3 log (X3i )+4 log (X4i )+5 log (X5i )+6 log (X6i )+ i (1) Imagine hypothetically that in da
Lecture 10: Dummy Variable Analysis
Econ 444, Winter 2010
Feb 17, 2010
Few observations to keep in mind about alternative functional forms : We only consider non linearity in variables (Y and Xs). The regression equation must be linear in parameters (s )
Lecture 11 : Multicollinearity
Econ 444, Winter 2010
Feb 22, 2010
Fig 1. No Correlation between independent variables X1 and X2 very rare as most economic variables are correlated with each other.
Fig 2 : Strong Correlation between independent variables X
Lecture 12 : Multicollinearity, Heteroskedasticity
Econ 444, Winter 2010
Feb 23, 2010
How to detect Multicollinearity ?
1
High R 2 but few signicant t-ratios may suggest presence of multicollinearity. High pairwise correlations among explanatory variables
Lecture 13: Heteroskedasticity
Econ 444, Winter 2010
March 1, 2010
3
Whites test : Consider following regression equation, Yi = 0 + 1 X1i + 2 X2i +
i
(1)
Suppose we suspect that the error term in the above equation is heteroskedastic. Whites test can be u
Lecture 14: Serial Correlation
Econ 444, Winter 2010
March 3, 2010
Serial Correlation : Motivation
When we have time series data we have information on same variable at different points in time. It is not surprising to nd that there is a high degree of co
Lecture 15 : Serial Correlation
Econ 444, Winter 2010
Mar 8, 2010
Testing for Negative Serial Correlation using DW test
H0 : = 0 no HA : < 0
serial
correlation serialcorrelation (1)
negative
To test for negative serial correlation we take the following st
Luke Porter
Econ 4400
Professor Ruddy
HW #3
1. y= 8.598+8.620x 1+0.0934x 2+7.260x 3+32.74x 4
2. y=8.598+8.620(0)+0.0934(12)+7.260(1)+32.74(0)
y=16.9788 drinks
So, the number of drinks per week for this person would be 17.
3. For an additional year/unit of