Practical statistics and data analysis
Notes for the statistics refresh lectures of the Nuclear and Sub-Nucleare Measurements Laboratory Course
Objectives
The goals of these lectures are:
- Refresh key concepts of probability and statistics
- Familiarize with common numerical methods and software tools for data analysis in nuclear and particle physics
Quite often physicists learn the basics of statistics in their very first year of classes, and yet they may come to appreciate what really means to “fit” some data with a certain model (or even what “model” really means) only years later. I find myself periodically revising the basics of statistics and data analysis, hopefully you will find this useful too.
All exercises and examples are available in this GitHub repository.
Content
- Lecture 0: Data analysis bootcamp: concepts, philosophy and tools
References
Probability and Statistics
- Statistical Data Analysis, Glen Cowan, Oxford University Press, 1998
- Statistics and Probability, Glen Cowan, PDG Review of Particle Physics Statistics - Probability
- Practical Statistics for the LHC, Kyle Cranmer, arXiv:1503:07622 [physics.data-an]
- Asymptotic formulae for likelihood-based tests of new physics, Glen Cowan et al., arXiv:1007.1727 [physics.data-an]
Tools
- ROOT, https://root.cern/
- git, https://git-scm.com/
- HSF: HEP Software Foundation Training Center, https://hepsoftwarefoundation.org/