Indiana University Bloomington

Economics E671
Econometrics Ill - Nonlinear And Simultaneous Models

Contact: (jescanci@indiana.edu) 10/24/2012
Offered: Spring, 2016
Capacity: About 20
Sequence: It is the third econometrics Ph.D. core course, after E571 AND E572.
Pre-Requisites: E571 (basic statistics and probability concepts) and E572 (linear regression)
Algebra Required: YES
Calculus Required: YES
Instructor: Juan Carlos Escanciano.
Days Per Week Offered: 2 days a week, One-two days computer lab in the course.
Recommended follow-up classes: E672 and Former E673
Syllabus: No Syllabus Avaliable
Keywords: Linear Simultaneous Models, Panel Data, Nonlinear Models, Probit and Logit, Maximum likelihood, Least squares, Generalized methods of moments, Monte Carlo, Bootstrap.
Description: The course aims to provide a solid theoretical background in econometrics for Ph.D. students. The prerequisites for the course are E571 and E572. Emphasis is on a thorough understanding of basic concepts and topics of modern econometrics. The core estimation and inference methods are maximum likelihood, (nonlinear) least squares and generalized methods of moments. All these estimators are studied in a unified way using the theory of extremum (M-) estimators. General approaches for econometric analyses of nonlinear models and with methods of econometric inference regarding the parameters of such models is essential for understanding applied articles in mainstream economic journals. The general theory will be illustrated with some specific examples that will be worked out using the statistical package Matlab. The student?s understanding of the course material will be tested using theoretical and practical data analysis assignments.
Books: Lecture notes are used (160 pages). No book required, but Wooldridge, J.M. (2002). Econometric Analysis of Cross Section and Panel Data. MIT Press, is recommended.
Substantive Orientation: Economics
Statistical Orientation: experimental
Applied/Theoretical: Theoretical course, but practical and applied aspects are discussed (e.g. how to implement estimators in a computer, use these estimators in applications, etc).
Software: None
How Software is Used: programming and data analysis
Problem Sets: YES, 5 ASSINGMENTS, 25% of final grade.
Data Analysis: At least 3 of the 5 assingments require data analysis
Presentation: No
Exams: yes, one midterm (25%) and a final exam (50%).
Comments: