e-Learning Support
Seminar: 03-M-AC-3 Semiparametric Statistics - Details

Seminar: 03-M-AC-3 Semiparametric Statistics - Details

Sie sind nicht in Stud.IP angemeldet.

Allgemeine Informationen

Veranstaltungsname Seminar: 03-M-AC-3 Semiparametric Statistics
Untertitel
Veranstaltungsnummer 03-M-AC-3
Semester SoSe 2026
Aktuelle Anzahl der Teilnehmenden 8
Heimat-Einrichtung Mathematik
Veranstaltungstyp Seminar in der Kategorie Lehre
Art/Form
Englischsprachige Veranstaltung Ja
ECTS-Punkte 4,5 / 6

Räume und Zeiten

Modulzuordnungen

Kommentar/Beschreibung

Statistical problems are described by statistical models. This means interpreting the data as realizations of random variables whose unconditional or conditional densities are described and estimated by statistical (regression) models. These models are usually identified by a set of parameters, which can be finite but also infinite dimensional. For this purpose, there are usually three types of possible models, depending on the structure of the data and the problem at hand: parametric, nonparametric, and semiparametric. A semiparametric model is characterized by the inclusion of both finite dimensional parametric and infinite dimensional nonparametric components. The main interest is usually in the finite dimensional parametric component, with the infinite dimensional component being co-estimated for the purpose of statistical inference and efficiency. In this seminar we will study the definition, properties, and applications of semiparametric models. Examples of semiparametric models include single-index models and Cox regression models for censored survival time data. We will also consider approaches to dealing with missing information in data sets. The use of semiparametric models plays a major role for medical studies, for example.

Prerequisites for participation in the seminar are basic knowledge of mathematical statistics (e.g. from Statistics 1) and of regression models (e.g. from Statistics 2). English speaking students are welcome.

In order to gain a good insight into the extensive theory of semi-parametric models, we will use the master thesis by Karel Vermeulen as our primary literature and study the chapters that are important for us in more detail, presenting the knowledge gained through the thesis in the form of individual presentations.

A list with the name of the thesis and further literature can be found below.

- Master thesis of Karel Vermeulen. Semiparametric Efficiency
- A. W. van der Vaart. Asymptotic Statistics. Cambridge Series in Statistical and Probabilistic Mathematics. Cambridge University Press, 1998.
- A. W. van der Vaart. ”On Differentiable Functionals“. In: Ann. Statist. 19.1 (März 1991), S. 178–204.
- Tsiatis, Anastasios. Semiparametric Theory and Missing Data. Vereinigtes Königreich, Springer New York, 2010.

Anmeldemodus

Die Auswahl der Teilnehmenden wird nach der Eintragung manuell vorgenommen.

Nutzer/-innen, die sich für diese Veranstaltung eintragen möchten, erhalten nähere Hinweise und können sich dann noch gegen eine Teilnahme entscheiden.