Willkommen beim Wiki "Statistik", WS 2014/15

Contact
Heike Siebert
siebert@mi.fu-berlin.de
Arnimallee 7, Raum 103
Office hours (during lecture time): Tuesday, 11:30-12:30

Bernhard Renard
RenardB@rki.de
Robert Koch-Institut, Nordufer 20, Raum 312
Office hours: after the lecture and by email appointment

Alexandra Reimers
alexandra.reimers@fu-berlin.de
Arnimallee 6, Raum 101

Therese Lorenz
therese.lorenz@fu-berlin.de
Arnimallee 6, Raum 102

News

To look at your graded repeat exam, please come to Heike Siebert's office (Arnimallee 7, Raum 103) on Wednesday, April 29, 14:00.

The results of the repeat exam are online.

The results of the final exam are online.

To look at your graded exam, please come to Heike Siebert's office (Arnimallee 7, Raum 103) on Thursday, Feb 19th, 15.00h

Repeat Exam date: Tuesday, April 14, 2015, 16-18, in HS 001, Arnimallee 3.

Beginning October 28 the lecture starts at 16:00.

Exam date: February 17 2015, 12-14, in HS 001, Arnimallee 3.

Please sign up on Campus Management for the class and the exam.

General Informationen

Lecturers: Bernhard Renard, Heike Siebert
SWS: 2
Exercises: Alexandra Reimers, Therese Lorenz
SWS: 2

ECTS: 6
Language: English/German

Dates and Locations

Lecture:
Tuesday 16-18h at Arnimallee 6, SR 032/A6
First lecture 14.10.

Exercises:
Wednesday 12:15 - 13:45, Arnimallee 6, SR 025/026
Wednesday 14:15 - 15:45, Arnimallee 6, SR 032

Topics

H. Siebert:
Mathematical background for Markov chains and related topics.
B. Renard:
Computational Statistics and Statistical Learning

Requirements

Exercises are mandatory, problem sets will be posted on this website on a weekly basis and are to be handed in at the Tuesday lecture. At least 50% of all graded problems need to be passed for a successfull participation. Attending the lecture is highly recommended. A 90 minute final examination determines the final grade.

Literature

Volker Schmidt. Markov Chains and Monte-Carlo Simulation, Lecture Notes University Ulm, 2010. Available here.
Pierre Bremaud. Markov Chains, Gibbs Fields, Mote Carlo Simulation, and Queues. Springer 1999. Available here.
Ehrhard Behrends, Introduction to Markov Chains (with Special Emphasis on Rapid Mixing), Vieweg, 1999.

Hastie, Tibshirani & Friedman. The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Springer, 2009. http://www-stat.stanford.edu/~tibs/ElemStatLearn/download.html

Lecture Materials

Part 1
The provided lecture notes do not constitute a complete script. Proofs, examples, remarks etc. presented in the lecture might be cut in part or even completely. However, all important definitions and theorems can be found in the notes.

01 Introduction and basic definitions (notes)

02 Canonical representation and n-step transition (notes)

03 Communication and periodicity (notes see above)

04 Recurrence and transience, absorption (notes)

05 Absorption, ergodicity (notes)

06 Ergodicity, reversibility (notes)

07 Markov Chain Monte Carlo, Metropolis-Hastings Algorithm (notes)

Part 2
The provided slides do not constitute a complete script. Proofs, examples, remarks etc. presented in the lecture on the board may be missing. However, all important definitions and theorems can be found in the slides. Further, additional reading material is provided which can help understanding the topics from a different perspective. The additional notes may exceed the material presented in the lecture (Only what was covered in the lecture is part of the final exam).

01 Introduction (slides)

02 Non-parametric tests (slides)(notes)

03 Normalization (slides)(notes)

04 Kernel Density Estimation (slides)(notes)

05 Nonparametric Regression (slides)(notes)

06 Support Vector Machines (slides)(notes)

07 Model Evaluation (slides)(notes)

08 Bagging and Random Forests (slides)(notes)

09 Boosting (slides)(notes)

Exercises

Problem sheet 1

Problem sheet 2
Skip the question concerning the cyclic classes in 7C.

Problem sheet 3

Problem sheet 4

Problem sheet 5 (Solution problem 18)

Problem sheet 6