ECTS - University of Pisa
The objective of the course is to provide students with a thorough coverage
of the classical econometric theory and with the computational tools to be
used in the empirical analyses. The emphasis is on the practical use of basic
econometric techniques to estimate economic relationships.
The course will focus on modern econometric techniques, addressing both
technical derivations and practical applications. Applications in the areas of
microeconomics and macroeconomics will be considered. By the end of the
course, students should have a good understanding of the fundamentals of
econometric theory and a critical understanding of applications of economet-
ric methods to empirical problems.
Students should be familiar with basic concepts in probability and statistics,
as well as linear algebra. The course includes a brief statistics and probability
refresher just in case. However, if you are not familiar with this material,
you should make time out of class to review it in detail.
Important: If you have not taken an introductory econometrics course,
preparatory reading is strongly advised, for example:
Gujarati, D.: Basic Econometrics. New York, McGraw-Hill, 2004.
1. Interpolation with Ordinary Least Squares Method (OLS)
2. Simple and K-variables Linear Regression Model
Basic assumptions , OLS estimation.
Algebraic Properties of the estimates, statistical properties of the estimates and the Gauss-Markov theorem.
The Coefficient of determination R2.
Unbiased estimation of sigma2.
The normality assumption and distributions of quadratic forms (no
proof).
t-test and F-test for testing linear hypothesis (linear restrictions).
3. Further results on the regression model
Functional forms: point elasticity, arch elasticity and semielasticity.
Dichotomous variables (dummy variables), structural changes and multicollinearity.
Restricted Least Squares (RLS).
Adding or deleting variables.
4. Generalized Least Squares (GLS)
Non spherical disturbances and OLS estimates, Generalized Least Squares (GLS) and Feasible Generalized Least Squares (FGLS).
Equivalence between GLS and OLS on transformed variables.
Eteroschedasticity (Estimation and White’s Test).
Autocorrelation.
6. Endogeneity
Endogenous regressors and inconsistency of OLS estimation.
Instrumental Variables (IV) and Two Stage Least Squares (TSLS).
Control Function (CF) approach: test and estimate.
7. Introduction to linear simultaneous equations models
Structural form and reduced form, simultaneous equations models and inconsistency of OLS estimation. Indirect Least Squares.
The identification problem.
Two Stage Least Squares (TSLS).
8. Introduction to Panel Data (dott. Giovanni Millo)
The main references for this courses are:
We will also provide students with some Handouts.
Software and Programming
Some lectures will be devoted to empirical applications and will require the
use of a statistical software. R is the statistical software for this course.
Students are not required to have any knowledge of R or other programming
experience, but they must be willing to learn.
For some applications also Matlab and possibly Gretl will be used.
There will be a midterm homework and a concluding written exam. The homework will be assigned at the beginning of the last week of lectures.
The homework has to be done in couple of students and returned in one week; both partners will receive the same grade on their homework.
The homework is compulsory and your final grade will be calculated as follows:
Midterm 30%
Concluding exam 70%
Total 100%
If necessary it will be possible complete orally the written test result.
