Course 2018-2019 a.y.

20532 - MACROECONOMETRICS

Department of Economics

Course taught in English
Go to class group/s: 31
GIO (6 credits - I sem. - OP  |  SECS-P/05)
Course Director:
LUCA SALA

Classes: 31 (I sem.)
Instructors:
Class 31: LUCA SALA


Prerequisites

Students attending this course should be familiar with basic concepts in Statistics and Linear Algebra. The topics covered in this course make use of and build on the following: Statistics: Random variable and random vector. Joint, marginal, and conditional distribution. Covariance and correlation. Conditional expectations and variances. Law of large numbers and central limit theorem. Estimator and estimate. Properties of estimators in finite and infinite samples (unbiasedness, consistency, efficiency, asymptotic normality). Least squares, method of moments, maximum likelihood. Hypothesis testing. Linear regression model. Linear algebra: Matrices. Linear independence. Rank and determinant. Inverse of a matrix. Eigenvalues and eigenvectors.

Mission & Content Summary

MISSION

The empirical analysis of macroeconomic data revolves on the study of time-series data. This course discusses thoroughly the statistical underpinnings of time-series analysis and shows how to apply those concepts to the analysis of the macroeconomy. The course also focuses on the important concept of identification, namely, on how to uncover causal and structural relationships populating economic models but hidden in the data. The course also discusses the most important applications in the literature. In so doing, students should replicate published papers. In the course, students also learn how to program using the software Matlab.

CONTENT SUMMARY

  • Stationarity.
  • Review of ARMA models. Specification and estimation of ARMA.
  • Non-invertibilities.
  • Non-stationarity.
  • Difference stationary vs Trend stationarity.
  • Testing for the presence of unit roots: the Dickey-Fuller test.
  • Spurious regression.
  • Simultaneous equation bias. The problem of identification.
  • The Sims’ critique to old macroeconometrics.
  • VAR models.
  • Granger causality (application: Sims, 1972).
  • Structural VAR and identification (applications: Sims, 1980, Blanchard-Quah, 1989 and Gali, 1999 news shocks and non-invertibilities).
  • Cointegration (application: King, Plosser, Stock and Watson, 1991).
  • Stationarity.
  • Review of ARMA models. Specification and estimation of ARMA.
  • Non-invertibilities.
  • Non-stationarity. Difference stationary vs Trend stationarity.
  • Testing for the presence of unit roots: the Dickey-Fuller test.
  • Spurious regression.
  • Simultaneous equation bias. The problem of identification.
  • The Sims’ critique to old macroeconometrics.
  • VAR models.
  • Granger causality (application: Sims, 1972).
  • Structural VAR and identification (applications: Sims, 1980, Blanchard-Quah, 1989 and Gali, 1999, news shocks and non-invertibilities).
  • Cointegration (application: King, Plosser, Stock and Watson, 1991).

Intended Learning Outcomes (ILO)

KNOWLEDGE AND UNDERSTANDING

At the end of the course student will be able to...
  • Be familiar with the main concepts and tools of time series analysis and being able to use them in other contexts.
  • Understand a vast majority of the scientific literature on time-series and macroeconometrics.
  • Identify what are the modelling assumptions underlying any structural macroeconometric model.
  • Translate the main assumptions in economic theories into restrictions on the empirical statistical model.

APPLYING KNOWLEDGE AND UNDERSTANDING

At the end of the course student will be able to...
  • Perform empirical analysis to uncover the effects of shocks in the economy.
  • Design a well-functioning VAR forecasting model.
  • Communicate effectively the empirical results of his/her analysis.
  • Use a well-known programming software, Matlab, to perform different kind of time-series analyses.
  • Do empirical analysis in a constructive way and think critically.

Teaching methods

  • Face-to-face lectures
  • Exercises (exercises, database, software etc.)
  • Group assignments

DETAILS

The learning experience of this course includes, in addition to face-to-face lectures, a number of classes in the Computer Laboratory, where the software Matlab is introduced. Students hand in 4 problem sets to be solved in groupwork. Problem Sets consist in replicating seminal papers in the literature of Structural VAR. The solution of the Problem Sets is discussed in the Computer Laboratory, where codes and results are shared. Students are encouraged to bring their own views and to share their insights.


Assessment methods

  Continuous assessment Partial exams General exam
  • Written individual exam (traditional/online)
    x

ATTENDING STUDENTS

To the end of measuring the acquisition of the above-mentioned learning outcomes, the students’ assessment is based on a final written exam. The exam consists of a mix of open questions and applied exercises. Attending students can deliver 4 problem sets. Problem sets teach students the use of Matlab to perform empirical analysis. Successful completion of the problem sets deliver up to 40% of the final grade. The remaining 60% are contributed by the final exam. Alternatively, students who do not wish to hand in the 4 problem sets can take a final written exam (general) that accounts for 100% of the final grad.


NOT ATTENDING STUDENTS

The assessment of non-attending students  follows the same rules as the assessment of attending students who do not hand in problem sets: 100% of the grade is set by the performance in the final exam.


Teaching materials


ATTENDING AND NOT ATTENDING STUDENTS

The main course material for both attending and non-attending students is:

  • L. SALA, Lecture note on Time Series Analysis.
  • W. ENDERS, Applied Econometric Time Series, last edition (selected chapters).
  • J.D. HAMILTON, Time Series Analysis, Princeton University Press, 1994 (selected chapters).
  •  Additional references are suggested during the course.
Last change 10/06/2018 18:14