Advanced Econometrics JEM005 : Course Syllabus

Transcription

1 Advanced Econometrics JEM005 : Course Syllabus Fall 2013/2014 Course Supervisor: Jozef Baruník ( barunik at fsv.cuni.cz ) Lecturer: Jozef Baruník Lectures: Wednesday 2:00 PM - 3:20 PM, room 109 ( and exceptionally 8:00-9:20 AM) Teaching Fellows: Tomáš Křehlík (TK) ( tomas.krehlik at gmail.com ) Krenar Avdulaj (KA) ( krenar.avdulaj at gmail.com ) Ayaz Zeynalov (JK) ( azerayaz at gmail.com ) Lucie Kraicová (LK) ( Kraicova.L at seznam.cz ) Seminars: Monday 5:00 PM - 6:20 PM, room 016 (TK, KA, JK, LK) Monday 6:30 PM - 7:50 PM, room 016 (TK, KA, JK, LK) Tuesday 2:00 PM - 3:20 PM, room 016 (TK, KA, JK, LK) Tuesday 3:30 PM - 4:50 PM, room 016 (TK, KA, JK, LK) Textbooks: For the topics covered during the semester, we will use chapters mainly from the two textbooks: (G) Greene, W.H. Econometric Analysis, 7th edition, Prentice Hall, 2012 please note that older versions of the text are fine too, just be aware of different chapters numbering (W) Wooldridge, J. Econometric Analysis of Cross-Section and Panel Data, Boston: MIT Press, 2010, 2nd edition In addition, following textbook can be used (CT) Cameron, C. and Trivedi, P.K. Microeconomertics: Methods and Applications, Cambridge University Press, New York, May 2005 Other good references: Maddala, Limited-dependent and Qualitative Variables in Econometrics, Cambridge, Davidson and MacKinnon: Econometric Theory and Methods, Oxford, Hayashi, F., Econometrics, Princeton: Princeton Univ. Press, Gourieroux and Monfort: Statistics and Econometric Models: Vol 1 and 2 (Cambridge University Press), 1995 (MHH) Martin, Hurn and Harris: Econometric Modelling with Time Series: Specification, Estimation and Testing, Cambridge University Press, 2013 Hansen, Bruce E.: Econometrics (online draft of graduate textbook), University of Wisconsin (last revision: January 2013) The Davidson and MacKinnon text is a good up-to-date text. Maddala s text, and the Wooldridge text are excellent for limited dependent and qualitative variables. Hayashi (2002) is much more involved with time series econometrics. Gourieroux and Monfort (1995) is an excellent complementary book for students interested in more technical text (at PhD level). Verbeek is less technical and undergraduate level. A new and rather broad text is Martin, Hurn and Harris (2013).

2 Software: The course statistical software is R (please see zero seminar on introduction). Course description and Objectives: The objective of the course is to help students understand several important modern techniques in econometrics and apply them in empirical research and practical applications. Emphasis of the course will be placed on understanding the essentials underlying the core techniques, and developing the ability to relate the methods to important issues faced by a practicioner. By completing this course, students will be able to use a computer based statistical software to analyze the data, choose appropriate models and estimators for given economic application, understand and interpret the results in detail (diagnose problems, understand proper inference) and will be confident to carry out the analysis and conclusions with respect to appropriatness and limitation of the methodology used. Finally, students will have sufficient grounding in econometric theory to begin advanced work in the field. Course pre-requisites: Basic Econometrics covered by J.M. Wooldridge, Introductory Econometrics, South-Western College Publishing (Econometrics I and Econometrics II courses) Assignments and Short Research Paper There will be three home assignments and one empirical project. Students are encouraged to form a groups of two or three students for solving the assignments as well as empirical project. In that case, please hand in only single solution with the names of both participating students. The detailed assignments will be posted on the web site of the course during the semester. Please hand in homework assignments as well as empirical project before class on the day they are due in the letter box of the corresponding teacher in front of the secretariat in a hard copy. Assignments handed in after the deadline will recieve no credit. Please append the log files to your assignments as well. In a short research paper, the group is expected to actively think of an economic problem to model, find the data and use the method of estimation covered during the semester. Research paper should be no longer than 6-10 A4 pages and should include short introduction to the problem, description of the dataset, method of estimation and finally the results. Main objective of carrying out the empirical excercise is to prepare students for work with real-world estimation problems and will complete the coursework. Students are encouraged to research the topic of interest, and replicate existing works with new data as well. Grading Assignments: 0-15% Midterm Exam: 0-20% Final Exam: 0-50% (to pass the Final, one needs to have at least 60% of correct answers) Empirical Paper: 15% 80% - 100% Grade 1 70% - 79% Grade 2 60% - 69% Grade 3 0% - 59% Fail

3 Schedule (Tentative): Week no. Date Topic Notes Assign.* Seminar Leader Seminar no. 1 Oct 2 Linear Regression TK Introduction to Estimation Frameworks in econometrics, Quantile Regressions TK Maximum Likelihood estimators TK Generalized Method of Moments 5 30 Simulation-based estimation and inference Lecture at 8:00 HW1 TK 3 KA 4 6 Nov 6 Midterm Exam no seminar 7 13 Endogeneity and Instrumental variables 8 20 Generalized Least Squares, non - i.i.d. errors KA 5 HW2 KA Models for Panel Data I AZ 8 10 Dec 4 Models for Panel Data II Lecture at 8:00 AZ Discrete Choice models HW3 AZ Extended Discrete Choice models Project LK Jan 8 LK 12 TBA Final Exams * the date when the assignment will be announced

4 Detailed course contents: 1. Linear Regression Revision using matrix algebra, finite sample properties, large-sample properties Reading: G(3-5: ), W (4: 49-76) 2-3. (2.1.) Introduction to Estimation Frameworks in econometrics Parametric estimation and inference (likelihood-based methods), semiparametric estimation (GMM, empirical likelihood), properties of estimators Reading: G(12: ) (2.2.) Quantile Regressions Quantile regressions, Quantiles and conditional quantiles Reading: G(7.3: ) and CT(4.6.) (2.3.) Maximum Likelihood estimators Basic likelihood concepts, score functions, principle of ML and its properties, Quasi and pseudo-mle Reading: G(14: ), W(13: ), or alternatively CT(5: ), MHH(1,2: 1:82 and 9: for QMLE) 4. Generalized Method of Moments The method of moments, GMM, properties, testing hypothesis in the GMM framework Reading: G( : ), W(14: ) or alternatively CT(6: ), MHH(10:361:396) 5. Simulation-based estimation and inference computer-intensive, simulation-based methods, bootstrap, maximum simulated likelihood estimation, moment-based simulation estimation Reading: G(15: ) or alternatively CT (11-13: , selection) or MHH(12: ) 6. MIDTERM 7. Endogeneity and Instrumental variables IV estimation, Multiple Instruments (2SLS), asymptotic theory and robust inference, measurement errors and omitted variables, Reading: G( : ), W (5: ) + Endogeneity in Systems of Equations (G : ?) if time allows 8. Generalized Least Squares, non - i.i.d. errors Generalized regression models and heteroscedasticity (efficient estimation via (F)GLS), Seemingly unrelated regressions Reading: G( : ), G( : ), W(7: )

5 9. Models for Panel Data I (static panel data methods) advantages of panel data; basics of linear panel models; pooled, random effects and fixed effect models; SUR versus Panel Data Models; target parameters and estimation by GLS; applications. Reading: G(11: ), W (10: ) or additional CT(21: ) 10. Models for Panel Data II (Dynamic linear paneldata models) Extensions of basic models; types of exogeneity; endogenous regressors; dynamic models; Discrete Choice Panel data methods, GMM methods for Panel models; Reading: G(11: ), G( : pp493) GMM in panel data, W (11: ) or additional CT(22: ) 11. Discrete Choice models Review of linear probability models for binary Discrete choice models, advantages, Logit and Probit models, specification issues Reading: W ( : ), G( : ) or additiona CT (14: selected) 12. Extended Discrete Choice models Multinomial logit and conditional logit models, Pooled discrete choice models Reading: W ( : ), G( : ) : 769:829 (only selection IF TIME ALLOWS) additiona CT (14, 23: selected)