In the seminar each participant will in turn pick a statistical paper that he likes and thinks is important and relevant. These can be for example some classical papers, review papers, tutorials or recent publications. The chosen paper will be presented then by the participant after having it distributed first to the all other participants so that they can read it before the meeting where the paper will be jointly discussed.

Seminar can be a part of Advanced Studies in Statistics (MTTS1 Other course in Mathematics or Statistics (advanced))

In the seminar each participant will in turn pick a statistical paper that he likes and thinks is important and relevant. These can be for example some classical papers, review papers, tutorials or recent publications. The chosen paper will be presented then by the participant after having it distributed first to the all other participants so that they can read it before the meeting where the paper will be jointly discussed.

Seminar can be a part of Advanced Studies in Statistics (MTTS1 Other course in Mathematics or Statistics (advanced))

In the seminar each participant will in turn pick a statistical paper that he likes and thinks is important and relevant. These can be for example some classical papers, review papers, tutorials or recent publications. The chosen paper will be presented then by the participant after having it distributed first to the all other participants so that they can read it before the meeting where the paper will be jointly discussed.

Seminar can be a part of Advanced Studies in Statistics (MTTS1 Other course in Mathematics or Statistics (advanced))

Contents of the seminar:

Biometric data sources: fingerprints, face and iris images. Data Preprocessing, data mining which includes analysis, classification and clustering, data evaluation and interpretation. Simple classification methods: K-NN, Linear discriminate, naive Bayes rule etc. Four Rates (TPR, FNR, TNR and FNR) and Equal error rate in biometric statistic.

Modes of Study:  Compulsorily attending seminar sessions and individual or group presentation.

This seminar is accepted as advanced studies in

• Computational Methods and Programming (laskentamenetelmien ja ohjelmoinnin opintosuunta) and
• Advanced courses in Algorithmics (algoritmiikan maisteriohjelma)
• Advanced Studies in Statistics

In the seminar each participant will in turn pick a statistical paper that he likes and thinks is important and relevant. These can be for example some classical papers, review papers, tutorials or recent publications. The chosen paper will be presented then by the participant after having it distributed first to the all other participants so that they can read it before the meeting where the paper will be jointly discussed.

Seminar can be a part of Advanced Studies in Statistics (MTTS1 Other course in Mathematics or Statistics (advanced))

The independent component (IC) model is a semi-parametric multivariate model where the observable observations are considered to be linear mixtures of unobservable latent variables which have independent components. The goal of independent component analysis (ICA) is to estimate the latent variables.

In this course we will discuss the IC model and its properties as well as introduce several ICA methods.

Related models and methods will also be shortly discussed.

Requirements: Students are expected to have a basic knowledge of multivariate methods and R.

Please fill the form and enroll before 7.3.2013. See the link below.

No classes on week 10, nor 1.-2.4.

Course is implementation of MTTS1 Other course in Mathematics or Statistics (advanced) and can be a part of Advanced studies in Statistics.

Students of Computer Science can include course in
- M.Sc. Programme in Algorithmics and
- Specialization in Computational Methods and Programming.

Contents of the seminar:

Biometric data sources: fingerprints, face and iris images. Data Preprocessing, data mining which includes analysis, classification and clustering, data evaluation and interpretation. Simple classification methods: K-NN, Linear discriminate, naive Bayes rule etc. Four Rates (TPR, FNR, TNR and FNR) and Equal error rate in biometric statistic.

Modes of Study:  Compulsorily attending seminar sessions and individual or group presentation.

This seminar is accepted as advanced studies in

• Computational Methods and Programming (laskentamenetelmien ja ohjelmoinnin opintosuunta) and
• Advanced courses in Algorithmics (algoritmiikan maisteriohjelma)
• Advanced Studies in Statistics

Short course by Erasmus visitor Augustyn Markiewicz (Poznan University of Life Sciences, PL).

Modes of Study

Lectures and lecture diary (more details later)

Topics covered

• Introduction of the theory of block designs
• Presentation of linear interference models 
• Derivation of the best linear unbiased estimators
• Study of statistical properties of the best linear unbiased estimators 
• Introduction to the theory of design's optimality 
• Comparison of optimal design under several interference models
• Block designs and their properties 
• The best linear unbiased estimation of treatments effects
• Properties of the best linear unbiased estimators
• Characterization of optimal design under several linear interference models

By email to Simo Puntanen (simo.puntanen@uta.fi).

Course is implementation of MTTS1 Other course in Mathematics or Statistics (advanced) and can be a part of Advanced studies in Statistics.