CEU, Economics Department

Lectures:  Gábor Kőrösi
Computer sessions: Balázs Muraközy
Course: 4 credits

Course Description

The subject is organized around selected economic problems. Each economic problem
has already been extensively analysed by econometric techniques. Lectures will review
some important empirical studies of the topic, explaining the way the economic problem
was represented by an econometric model. The course will concentrate on the practical
use of econometric methods, reviewing the relevant methodology, its use, and the possible
alternative modelling approaches. The lectures are supplemented by computer classes,
where students can gain hands-on experience in applied econometric analysis.

Text for an important part of the course:

Berndt, E. R. [1991]: The Practice of Econometrics: Classic and Contemporary;

Recommended readings for prerequisites, revision and additional theory:

Greene, W. H. [2003] or [2000]: Econometric Analysis; Prentice Hall, (5th or 4th Ed.).

Korosi G., L. Matyas and I. Szekely [1992]: Practical Econometrics; Avebury, Aldershot.

Maddala, G. S. [2001]: Introduction to Econometrics; Wiley (3rd Ed.).

Mills, T.C. [1999]: The Econometric Modelling of Financial Time Series; Cambridge
University Press

Course Outline:

Lecture 1: The demand for electricity

Econometric theme: Time series sample; error autoecorrelation; functional form.

Lecture 2: Predictability and causality

Econometric theme: Models of stationary time series: DL, AR, ARMA; Granger causality,

Lecture 3: Small simultaneous macromodels

Econometric theme: Simultaneity.

Lecture 4: Consumption (1): Short and long run, the error correction mechanism

Econometric theory: Specification analysis and model selection; error correction.

Lecture 5: Consumption (2): Reactions to Lucas' critique; from the Euler equation
approach to the excess sensitivity hypothesis

Econometric theory: Unit root

Lecture 6: Consumption (3): Cointegration

Econometric theory: Cointegration and error correction

Lecture 7: New Keynesian Phillips curves

Econometric theme: Expectations, GMM, VAR

Lecture 8: Returns to scale and price markups

Econometric theme: Panel data models.

Lecture 9: Labour demand

Econometric theme: Dynamic panel data models.

Lecture 10: Riskmetrics I: ARCH

Econometric theme: ARCH model family.

Lecture 11: Riskmetrics II: Transaction level data modelling

Econometric theme: Duration models, ACD.

Lecture 12: Modelling strategies

Conference presentations (group assignment)


For audit: Students should get at least 50% for the projects; exam is not required.

For grade: The final mark will be composed of two components: projects and examination.

A. Projects: Students will have to complete two empirical projects during the semester; one
group assignment and another individual one.

i.) Group assignment:
Randomly selected groups will have to do a joint empirical analysis, and to write a report on a
given topic. They will have to explore some macroeconomic processes in specific countries,
using econometric tools for inference. They will have to compare the behavioural patterns of
the analysed economies. The empirical results should be summarised in a report, and also
presented at the conference session on the practical classes last week. Reports should be
filed on 14 November 2005 (Monday), or before. Data used for the empirical work should also
be attached. Both the report and the dataset must be submitted in an electronic form. Each
group will have to discuss two reports at the conference session. 30 minutes will be allocated
to each report at the conference session: 15 min for presentation, 5-5 min for both discussants,
and 5 min for general discussion. Groups will be marked instead of individuals, i.e., each group
member will get the same mark for the joint work. Weights: 10% for the report, 5% for the
presentation, 5-5% for the two discussions. (Total: 25%)

ii.) Individual assignment:
Each student can choose any topic and suitable dataset for the project, except for the topic of
the group assignment. Reports should be filed on 30 November 2005 (Wednesday), or before.
The report must include:

The report should not exceed 10 double spaced pages (i.e., about 3000 words), although
appendices can be attached (e.g., printed output of estimation results). Be brief, be precise,
and do not waffle. Excessively long reports will be penalized. Reports should be typewritten.
Attach the dataset to your report on disk.


The individual assignment gives 25% of the final mark.

B. Examination: Its weight is 50% of the final mark.