Statistical Data Analysis 2

Course guarantor:

Lecturer:

Tutor:

Supervisor:

Department of Physics

Synopsis:

Individual students work will include implementation and testing of a program for analysis of generated data sample. Background understanding of Monte Carlo generators for hadron collision will be explained. The course covers methods of data smearing and subsequent deconvolution of data. Basics understanding and usage of neural networks and machine learning will be covered.

Requirements:

Syllabus of lectures:

1. Gaussian noise and detector response.

2. Fisher information.

3. Testing of hypotheses.

4. Maximum entropy method.

5. Deconvolution Bayes approach, SVD method a HBOM method.

6. Methods for generation of random samples.

7. Simulation of random processes via the Monte Carlo method.

8. Neural networks, large data samples.

Syllabus of tutorials:

Study Objective:

Study materials:

Key references:

[1] L. Lista, Statistical Methods for Data Analysis in Particle Physics, Springer, 2017.

[2] G. Cowan, Statistical Data Analysis, Clarendon Press, Oxford, 1998.

Recommended references:

[3] D. S. Sivia, Data Analysis A Bayesian Tutorial, Claredon Press, Oxford, 1998

[4] C. Maña, Probability and Statistics for Particle Physics, Springer, 2017.

Equipment:

Computer classroom, software ROOT

Note:

Further information:

No time-table has been prepared for this course

The course is a part of the following study plans:

• Jaderná a částicová fyzika (compulsory course in the program)
• Kvantové technologie (elective course)