LECTURE PLAN: ECON 4160 ECONOMETRICS
Modelling and systems estimation
Time and Place:
Thursday 10:15 – 12, auditorium 5
No lectures in weeks 12 and 13
First lecture: January 20
Language:
English
Lecturer:
Erling Barth
E-mail: [email protected]
Web-page: http://www.socialresearch.no/staff/ebarth
Course description and reading list:
See: www.uio.no/studier/emner/sv/oekonomi/ECON4160/
The course builds on ECON 3150/4150
Tentative lecture plan:
The following topics will be covered in the lectures.
The week numbers given in parenthesis for each topic are tentative.
Exercises and the compulsory paper will be given in the seminars.
The reading list is a suggestion from the lecturer and may cover parts of the
reading list for the course, only.
1. Introduction (week 3)
Basic characteristics of econometrics.
Survey of the course and remarks on the reading list.
Point of departure: Ordinary least square (OLS).
Properties of the OLS estimators.
Hypothesis testing
Biørn: Ch 1, 2.1-2.4, App. 2A and 2B
Greene Ch. 2-5.3, 7.1-7.4, 8.1, 8.2 App. A-C
Lecture Notes: Lecture note 1
I. SINGLE EQUATION MODELS
2. Generalized linear regression models (week 4)
The Generalized Least Squares (GLS) method.
Feasable Generalized Least Squares (FGLS).
Heteroskedastic and autocorrelated disturbances.
Biørn: Ch 3.1-3.4
Greene Ch. 10.1-10.3, 11.1-11.5, 12.1-12.3, 12.7, 12.9
3. Regression equations with simple dynamics (week 5)
Modelling delayed responses in economics
Distributed lags.
Finite and infinite lag distributions.
Greene Ch. 19.1, 19.2.1, 19.3.1
Lecture Notes: “Distributed lags”
4. Discrete dependent variables (week 6 & 7)
Linear probability model (LPM).
The binary choice models.
The binomial Logit and Probit models.
Greene Ch. 21.1-21.3
Lecture Notes: “Discrete Choice”
III) MULTI-EQUATION MODELS
5. System of linear regression equations (week 8-9)
Models with two regression equations.
Seemingly Unrelated Regressions (SUR).
Constraints across equations
Biørn: Ch. 4, 5, 6.3.a, 6.5
Greene Ch. 14.1, 14.2, 14.3
Lecture Notes: 2, 9
6. Limitations of regression analysis (week 10)
Why do we need simultaneous equation systems?
Answer 1: Endogenous r.h.s. variables.
Answer 2: Measurement error in exogenous variables.
Simultaneity problem and simultaneity bias.
The inconsistency of OLS.
Biørn: Ch 8.1, 10.1, 10.2
Greene Ch. 15.1-15.2
7. Model concepts and the identification problem (week 11 & 14)
The identification problem illustrated in simple market models. Is meaningful estimation at all possible?
The identification problem in measurement error models.
Basic concepts: Structural coefficients, structural form, reduced form. Normalization restrictions. Exclusion restrictions. Structural equations and autonomy. Confluent relations.
The identification problem as a general problem in econometrics.
The identification problem in simultaneous equation systems.
The order condition. “Overidentification”.
Biørn: Ch. 7
Greene: Ch. 15.3
Lecture Notes: 3, 4 ,5
8. Estimation and test methods for simultaneous equation systems. (week 15&16)
A basic concept: An instrument variable IV
Estimation of “exactly identified” and “overidentified” structural equations by means of IV’s.
Testing for instrument relevance and instrument exogeneity.
Using IV to estimate models with errors in variables.
Biørn: Ch. 8.2 10.3, 10.4
Greene: Ch. 15.4, 15.5.1 , 15.5.2
Lecture Notes: 6, 7
9. Two-Stage Least Squares methods (week 17 & 18)
Estimation of “overidentified” equations by means of Two-Stage Least Squares (2SLS).
Testing for overidentification.
Biørn: Ch. 8.3.-8.5, 10.3-10.4
Greene: Ch. 15.5.3-15.5
Lecture Notes: 8
III. SYNTHESIS.
10. A survey of econometric models (week 19)
Survey of single-equation models discussed in the course.
Survey of multi-equation models discussed in the course.
The need for structural relationship in empirical research.
Interdependent versus recursive simultaneous models.
Unidirectional versus multidirectional “causality chains”.
Greene: Ch. 15.2.3
2002 FRESHFIELDS LECTURE 19 ARB INT’L 279 (2003) ARBITRATION’S
3 TEST QUESTIONS – BIT LECTURE 1 OBJECTIVES AND
3012 FUNDAMENTALS OF MATERIALS SCIENCE FALL 2003 LECTURE 10
Tags: econometrics modelling, econometrics, estimation, lecture, systems, modelling