4585 Y1P4 Econometrics
ao.Univ.Prof. Dr. Alois Geyer
  • LV-Typ
  • Semesterstunden
  • Unterrichtssprache
05.02.2019 bis 24.02.2019
Anmeldung über LPIS
Hinweise zur LV
Planpunkt(e) Master
Wochentag Datum Uhrzeit Raum
Mittwoch 08.05.2019 14:00 - 15:30 TC.5.01
Mittwoch 08.05.2019 15:30 - 18:00 TC.4.01
Mittwoch 15.05.2019 14:15 - 15:45 D5.1.001
Donnerstag 16.05.2019 09:00 - 11:00 TC.5.27
Mittwoch 22.05.2019 14:15 - 15:45 D5.1.001
Donnerstag 23.05.2019 09:00 - 11:00 TC.3.03
Mittwoch 29.05.2019 14:15 - 15:45 D5.1.001
Mittwoch 05.06.2019 14:15 - 15:45 TC.2.03
Mittwoch 05.06.2019 16:00 - 18:00 TC.4.01
Donnerstag 06.06.2019 09:00 - 11:00 TC.4.01
Mittwoch 12.06.2019 14:15 - 15:45 D5.1.001
Donnerstag 13.06.2019 09:00 - 11:00 TC.4.01
Mittwoch 19.06.2019 14:15 - 15:45 D5.1.001
Freitag 28.06.2019 09:00 - 10:30 D3.0.225

Inhalte der LV

This course covers basic topics in econometrics with a focus on applications in finance.

Lernergebnisse (Learning Outcomes)

After completing this course participants will

  • have the ability to apply and interpret the results of regression analyses
  • be familiar with key aspects relevant for the specification of a regression model
  • understand the relevance and implications of various assumptions in each step of the analysis
  • know why and how specific properties of regression residuals must be tested
  • understand the consequences of violations of certain assumptions, and know how to account for them
  • be familiar with basic definitions of financial returns, and able to derive and interpret their empirical (dynamic) properties
  • know how to distinguish non-stationary from stationary series and how to apply unit-root tests
  • understand the purpose and the basic principles of GARCH models, and how to estimate and test such models

Regelung zur Anwesenheit

Participants are required to attend each class, except for serious illness and/or important private concerns.


The course is taught using a combination of lectures and practical examples demonstrated in class. The lectures are aimed at establishing a sound understanding of the main ideas and basic principles of econometric methods and analyses. Special emphasis is put on applications using financial data. For that purpose, practical examples will be presented in class. Data used in these examples are also available to participants and can be downloaded from This provides the opportunity to replicate the examples on their own, prepare for homework assignments and the final exam.

Preparation: Participants are expected to have read the appropriate sections from the lecture notes (required readings are specified below for each unit).

Participants have to do assignments based on the exercises specified in the lecture notes. The purpose of the assignments is to practice using actual data, to recall the methods' theoretical basis and assumptions, and to get acquainted with empirical evidence on financial data.

Participants have to engage in a prediction competition (Kaggle InClass). For that purpose, they have to develop a prediction model for a dataset which will be made available after the first unit. Predictions are evaluated on the basis of a holdout dataset. Participants will get feedback whenever they submit a prediction. The submission of predictions, feedback and monitoring are done via a specific website (details will be provided in the first unit). In each unit, the progress made in developing the prediction model will be discussed. I will provide feedback and make recommendations.

Leistung(en) für eine Beurteilung

  • 20% of the final grade is based on the final exam.
  • 20% of the final grade is based on homework assignments. Homework assignments can be done in groups consisting of up to 3 members. Each member of a group must be able to explain all aspects of an assignment (i.e. it is not recommended that group members only do parts of an assignment; all group members should work jointly on the assignment and must take full responsibility).
  • 30% of the final grade is based on the performance in short quizzes/tests done during class. These tests consist of multiple-choice questions, which are based on the topics actually covered in class. In principle, participants should be able to correctly answer the questions by actively following the presentations and discussions during class. Note that knowing formulas or commands of various software packages are not required for successfully answering the questions!
    Quizzes are executed using the "clicker" feature of Learn@WU. In order to participate in a quiz, each participant has to make sure she/he can access Learn@WU via a laptop, tablet, or smartphone.
  • 30% of the final grade is based on the score obtained in the prediction competition.
  • Participants may receive extra credits for valuable contributions made during class (e.g. when discussing homework assignments).




Teilnahmevoraussetzung(en) und Vergabe von Wartelistenplätzen

Successful completion of Mathematics I and Financial Markets and Instruments.


1 Autor/in: Alois Geyer

Anmerkungen: Lecture notes
Jahr: 2019
Prüfungsstoff: Ja
Empfehlung: Unbedingt notwendige Studienliteratur für alle Studierenden
Art: Skriptum

Empfohlene inhaltliche Vorkenntnisse

Participants should be familiar with the following topics on an undergraduate level:

  • Mathematics (e.g. matrix algebra, polynomials, derivatives, etc.)
  • Probability(e.g., distributions, conditional probability, expectation operators, etc.).
  • Statistics(e.g., descriptive statistics, sampling distributions, hypothesis testing,etc.)
  • Computing: Excel, EViews, R

Erreichbarkeit des/der Vortragenden

I answer e-mails as soon as possible. Meetings in my office can be arranged via


Section numbers for required readings refer to the lecture notes.

In each unit, one or several practical examples are used to demonstrate principles and methods introduced in class.

Lecture notes:

Associated files:

Detailinformationen zu einzelnen Lehrveranstaltungseinheiten

Einheit Datum Inhalte

Basics of regression analysis (sections 1.1 and 1.2)

Participants will become familiar with fundamental aspects of linear regression analysis. In particular, they will be able to use the appropriate terminology, will understand the principle of least squares, and will have a clear understanding about the relevance and implications of various assumptions in each step of the analysis. Participants will be able to interpret the results of a regression analysis, in particular, the coefficients of the equation, the goodness of fit indicators, and test statistics.


Specifications (section 1.6):

Participants will become familiar with key aspects of the specification of a regression model. In particular, they will know how and when to apply log transforms, dummy variables, and interaction terms. They will also know how to interpret the coefficients in each of these cases. In addition, they will know which consequences are associated with omitted and irrelevant regressors, they will become familiar with considerations and criteria in (the sequence of) selecting regressors, and they will learn about the role of cross-validation in the specification search.


Regression diagnostics and GLS (sections 1.7 and 1.8):

Participants will know which properties of regression residuals must be tested. In particular, they will know why non-normality, heteroscedasticity, and autocorrelation must be tested, how these features can be tested, and how these properties are related to the assumptions introduced in unit 1. Participants will know the consequences of violations of the required properties, and they will know how violations can be accounted for or corrected.


Financial time series (section 2.1):

Participants will become familiar with basic definitions and properties of simple and log returns, typical empirical features of financial returns, and frequently used distributions. Participants will also know how to use and interpret autocorrelation analyses to describe the dynamic properties of (absolute) returns. They will be able to define abnormal returns, and understand the basic idea and results of event studies.


ARMA models (section 2.2):

Participants will become familiar with definitions and properties of autoregressive (AR), moving average (MA) and ARMA models. They will know how to identify which of these models is appropriate on the basis of (partial) autocorrelations. Participants will know how the estimated coefficients and residuals can be used to determine the appropriateness of a model. They will be able to compute forecasts and will be able to relate properties of forecasts to autocorrelation patterns.


Non-stationary models (section 2.3):

Participants will know how to distinguish non-stationary from stationary series in terms of statistical features and forecasting behavior. They will understand the basic idea and special nature of unit-root tests, how the tests are carried out, and which conclusions about the underlying series can be derived.


GARCH models (section 2.5):

Participants will become familiar with the purpose and the basic principles of GARCH models. They will know how to specify the model equations (conditional mean and variance), and know how to estimate GARCH models using the maximum likelihood principle. Participants will know why and how GARCH models account for stylized facts of financial returns such as non-normality and volatility clustering. Participants will know which properties of residuals must be checked, and how variance forecasts can be computed.

Zuletzt bearbeitet: 14.04.2019