Statistical Data Analysis 2

Garant předmětu:

Lecturer:

Tutor:

Supervisor:

Department of Physics

Synopsis:

Individual work will include implementation and testing of a program for analysis of generated data sample. Results are reviewed during the exam

Requirements:

Knowledge of basic course of probability and statistics.

Syllabus of lectures:

1. . Gaussian noise.

2. Fisher information.

3. Model selection I.

4. Model selection II..

5. The principle of maximum entropy.

6. Goodness of fit. I.

7. Goodness of fit II.

8. Unfolding I. - Bayesian method.

9. Unfolding II. - SVD method.

10. Methods for generating random samples.

11. Monte Carlo method.

12. Neural networks, big data.

Syllabus of tutorials:

1. . Gaussian noise.

2. Fisher information.

3. Model selection I.

4. Model selection II..

5. The principle of maximum entropy.

6. Goodness of fit. I.

7. Goodness of fit II.

8. Unfolding I. - Bayesian method.

9. Unfolding II. - SVD method.

10. Methods for generating random samples.

11. Monte Carlo method.

12. Neural networks, big data.

Study Objective:

Knowledge:

advanced application of statistical methods for experimental data analysis, applicability of various methods, data filtering, testing of hypotheses

Skills:

orientation in the field, ability to analyse experimental dat

Study materials:

Compulsory literature:

D.S. Sivia ? Data Analysis ? A Bayesian Tutorial, Oxford, 2006.

F. James: Statistical methods in Experimental physics, World Scientific, 2006

Optional literature:

G. Cowan, Statistical Data Analysis, Clarendon Press, Oxford, 1998.

T. Eadie et al., Statistical Methods in Experimental Physics, Amsterdam,1971.

Note:

Further information:

No time-table has been prepared for this course

The course is a part of the following study plans: