Econometrics 2 (ECO 7425)
Class. No. 84380
Department of Economics, Florida International University
Fall Semester 2009

 

Lecture Hours: Tue & Thurs 14:00-15:15 

Syllabus
 

Sample Programs

Sample Regression Program

Sample NLS Program                          Data for NLS Sample Program

Sample GMM Program                        Data for GMM Sample Program

Sample MLE Program

 

Homework

Data for Homework 1

Data for Homework 2              Solution Program

Data for Homework 3

Data for Homework 4

 

Lecture 

Topics covered

 

1

GAUSS Program – Sample Regression Program

 

2

GAUSS Lab Session

 

3

I. Nonlinear Least Squares (NLS); Non-Linearity in Variables and Non-Linearity in Parameters; Examples of Non-Linear Models; Estimation – Non-Linear Least Squares

 

4

II. Numerical Optimization; Principles of Numerical Optimization; Univariate Search Techniques – Grid Search Method; Direct Search Methods – Simplex Method; Descent Methods – Method of Steepest Descent; Convergence Criteria; Numerical Evaluation of Derivatives

 

5

Newton-Raphson Method; Quasi-Newton methods; Constrained Optimization –Algebraic Transformations;

 

6

PDFs of Transformations of Random Variables; Standard Errors by the Delta Method

 

7

Properties of the NLS Estimator; GAUSS Program – Sample Optimization Program

 

8

Selection of Starting Values - Method of Moments; III. Method of Maximum Likelihood (MLE); The Principle of Maximum Likelihood; The Likelihood Equations; Examples – classical linear regression model, non-linear regression

 

9

Computational Aspects; The Cramer-Rao Lower Bound; Properties of MLE

 

10

IV. Test Procedures; Specification tests; Derivation of the Likelihood Ratio (LR) Test

 

11

The Lagrange Multiplier (LM or Score) Test, and the Wald Test (linear constraints); Wald Test (non-linear constraints); Monte Carlo Simulations

 

12

V. Generalized Method of Moments (GMM); Orthogonality Conditions implied by Economic Theory

 

13

GMM estimator; Optimal Distance (Weighting) Matrix; Computation of the GMM estimator; Distribution of the GMM estimator

 

14

GMM Program Discussion; Test for Over-Identifying Restrictions; HW 2 Solutions (NLS)

 

15

VI. Generalized Least Squares (GLS);  Non-Scalar Identity Covariance Matrix; Non-Spherical Disturbances; Properties of the OLS Estimator; Consequences of Using OLS Estimator; Heteroskedasticity-Consistent Covariance Matrix Estimation

 

16

The GLS Problem; The GLS Estimator; Properties of the GLS Estimator; Example: Covariance Matrix under Pure Heteroskedasticity; Weighed Least Squares (WLS) – a Special Case of GLS

 

17

Feasible GLS; Properties of Feasible GLS Estimator; Consistency of OLS  and Feasible GLS

 

18

The t- and F- test statistics (with and without the assumption of normality); The t- and F- test statistics under Feasible GLS

 

19

Heteroskedasticity and Cross-Correlation

 

20

VII. Seemingly Unrelated Regressions (SUR); Systems of Equations; Estimation

 

21

Project Discussion; Systems of Equations Special Cases - No Correlation across Errors, Identical Regressors across Equations; Equivalence of GLS and OLS; Testing for Correlation between Errors

 

22

Imposing Cross-Equation Restrictions; Combining Cross-Sectional and Time Series Data

 

23

VIII. Simultaneous Equations Models

 

24

Problem of Identification; Endogenous & Exogenous Variables; Structural Equations; Problems with OLS Estimator – Simultaneous Equations Bias; Reduced Form Equations; Estimation – Indirect Least Squares (ILS)

 

25

Non-Uniqueness of ILS; Instrumental Variable (IV) Estimation; Two-Stage Least Squares (2SLS)

 

27

Limited Information vs Full Information Estimation; Three-Stage Least Squares (3SLS); Conditions for Identification; Normalization

 

28

IX.  Time Series

 

29

Student Presentations

 

30

Student Presentations