You are here: Wiki>DiBiMath Web>StatistikWS17 (19 Apr 2017, BernhardRenard)Edit

Willkommen beim Wiki "Statistik", WS 2016/17

Contact

Heike Siebert
siebert@mi.fu-berlin.de
Arnimallee 7, Raum 103
Office hours: by email appointment

Bernhard Renard
RenardB@rki.de
RKI, Nordufer 20, Room N01.O2.014
Office hours: by email appointment

Jakob Schulze
jakob.schulze@fu-berlin.de

News

Here are the results of the repeat exam.
The repeat exam will be held at 19.04.2017, 16-18h In Taku 9/006. You can have a look at your graded exam ('Einsicht') on Thursday, 20.04.2017, 16.00h at Heike Siebert's office (Arnimallee 7, Room 103).

Here are the results of the final exam.
An R course is offered in an FU qualifying programm. More information here. There are also many free online courses such as this one. We also provide a small tutorial as exercise 0. Please also use the opportunity to ask in the exercises for help before the first real R problem is posted.
You can get extra credit for the exercises (if you have not received >50% of the points on the reviews) by handing in your solution to problem set 15.

Exam

A 90 minute final examination will be held at 16.02.2017, 12-14h in Taku 9/HS 028 (Großer Hörsaal). You are allowed to have a single page (DINA4, front side only) of hand written notes ‎at the exam.

The repeat exam will be held at 19.04.2017, 16-18h In Taku 9/006.

You can have a look at your graded exam ('Einsicht') on Friday, 17.02.2017, 9.30h at Heike Siebert's office (Arnimallee 7, Room 103).

General Informationen

Lecturers: Heike Siebert, Bernhard Renard
SWS: 2
Exercises:
SWS: 2
ECTS: 6
Language: English

Dates and Locations

Lecture:
Thursday 12-14h, Takustr. 9, SR 006/T9
First lecture 20.10.

Exercises:
Monday, 12-14h in SR 032 (Arnimallee 6), 14-16h, SR 025/026 (Arnimallee 6)

First exercises 24.10.

Topics

Mathematical background for Markov chains and related topics
Computational Statistics and Statistical Learning

Requirements

Exercises are mandatory. Problem sheets will be available every Thursday and discussed the following Monday in the exercises. Problem sheets should be thoroughly worked through, however, solutions need not be handed in. Three Problem Reviews (45 min) will be written during term. To achieve "active participation" in this class, 50% of the Review points are needed.
Review dates: 21.11.2016, 09.01.2017, 30.01.2017, 13:15-14:00 in SR 032, Arnimallee 6.
Attending the lecture is highly recommended. A 90 minute final examination on 16.02.2017 determines the final grade.
Both for the reviews and the exam no tools (script, calculator...) besides a pen are allowed. However, you are allowed to have a single page (DINA4, front side only) of hand written notes ‎at the exam. Please bring a student and a photo ID.

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.
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. Available here.

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 (notes)

06 Ergodicity, Reversibility (notes)

07 Markov Chain Monte Carlo (notes)

08 MCMC II (notes see above)

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-parametrics (slides) (reading material)

03 Kernel Density Estimation (slides) (reading material)

04 Kernel Regression (slides) (reading material)

05 Model Evaluation (slides) (reading material)

06 Support Vector Machines (slides) (reading material (German)) (alternative reading material (Chapters 4.3, 12.1-12.3))

07 Classification Trees (slides) (reading material)

08 Bagging and Random Forests (slides) (reading material)

09 Boosting (slides) (reading material)

10 Normalization (slides) (reading material)

Exercises

R Intro data

Problem sheet 1

Problem sheet 2

Problem sheet 3

Problem sheet 4

Problem sheet 5

Problem sheet 6

Problem sheet 7

Problem sheet 8

Problem sheet 9

Problem sheet 10 data

Problem sheet 11 data

Problem sheet 12 data data

Problem sheet 13 data

Problem sheet 14 data

Problem sheet 15 data

Topic revision: r33 - 19 Apr 2017, BernhardRenard
 
  • Printable version of this topic (p) Printable version of this topic (p)