SHORT COURSE: «On aspects of statistical modelling»

M.Sc. in Statistics

SHORT COURSE: «On aspects of statistical modelling»

Ioannis Kosmidis, Professor, Department of Statistics , University of Warwick 

Lecture 1

Wednesday

9 April 2025

609

(Evelpidon 47A & Lefkados 33,6th floor)

12.00-15.00

Lecture 2

Thursday

10 April 2025

609

(Evelpidon 47A & Lefkados 33,6th floor)

09.00-12.00

Lecture 3

Friday

11 April 2025

609

(Evelpidon 47A & Lefkados 33,6th floor)

12.00-15.00

Lecture 4

Saturday

12 April 2025

609

(Evelpidon 47A & Lefkados 33,6th floor)

09.00-12.00



 

The course is financed by the “M.Sc. in Statistics” of Athens University of Economics and Business.

As a limited number of positions is available, if you wish to attend the above short course, you must apply in this link until Friday March 28, 2025.

Course Description

Aim:

To introduce important aspects of statistical modelling, including model  selection, various extensions to generalised linear models, non-linear  models, and latent variable models, and the associated methods for  estimating and drawing inference from those.

Learning outcomes:

After taking this module, students should be able to:

* Provide a theoretical justification for the use of various criteria for model selection and apply these techniques in practice.

* Describe some reasons why Generalised Linear Models may fail to fit real data well and apply techniques to diagnose such failures.

* Describe some commonly used extensions to Generalised Linear Models, and conduct frequentist and Bayesian inference for these models.

* Identify and describe latent variable models and key algorithms for estimating them.

Prerequisites:

Most concepts will be re-introduced, but familiarity with linear and generalized linear models, and likelihood and Bayesian inference, will help.

Topics:

- Principles and practice of model selection.

- Extensions of the generalised linear model, including models for overdispersion and mixed-effects models.

- Non-linear models.

- Latent variable models.