Physics 250- Fall 2021
Statistical Methods for data analysis in (mostly) Particle Physics


Meeting coordinates: Wednesday 5:00-6:15 South Hall 1609
(occasionally also on Mondays, TBD)
http://hep.ucsb.edu/people/claudio/Phys250


Instructor:
Claudio Campagnari
Email:
claudio@physics.ucsb.edu
Office:
Broida 3001.  Phone 805-893-3742.
Office Hours:
By appointment, or just drop by


Administrative Assistant:
Debbie Ceder
Email:
dla@physics.ucsb.edu
Office:
Broida 5014.  Phone 805-893-2058

Note: the main door to my office is in the department administrative office suite which, because of the Covid situation, is usually locked to keep the traffic down.  You can use the other door: go right as you come out of the elevators on the 3rd floor of Broida, go through the glass doors, my other office door is immediately to the right, opposite the ladies room.  Just knock....

An important statement from the instructor and the Physics Department about our expectations for professional behavior and suggestions on how to deal with problems.


Modern Particle Physics experiments rely heavily on statistical analyses.  As the experiments and the analyses grow in complexity, so do the statistical methods and the tools that are applied.  Some of the tools have evolved into poorly documented complex software black boxes that are often used blindly.  This course is a practical introduction to statistics aimed at providing a foundation towards the understanding of the methods that are commonly applied. 

While the treatment is tailored to students in particle physics, many of the concepts could be of interest to students in other fields as well.  This will be a two-unit Pass/Fail graduate lecture course that will meet once a week (we will however reserve a second weekly lecture time that will be used only in case of need).  The course has some overlap with an existing course PHYS 240: Statistics, Data Analysis, and Machine Learning for Physicists, however PHYS 240 does not serve the needs of particle physics students very well.

Topics to be covered tentatively include:
Probability, probability distribution functions.  Bayesian vs. frequentist statistics.  Useful probability distributions: Poisson, Gaussian, Lognormal, Binomial, Beta, Gamma, Breit-Wigner, Cruyff.   Confidence/credible intervals.  Limits on number of events:  bayesian,  frequentist, Feldman-Cousins.  Hypothesis testing, type I and type II errors, p-values.  Include systematics.  Central limit theorem. Fitting: chi-squared, log-likelihood, extended log-likelihood.  Correlations.  How to use Minuit.  Toy MC.  Kolmogorov-Smirnov. Barlow-Beeston method. Nuisance parameters, profile-likelihood, CLs method. Saturated goodness of fit. sPlots (maybe).

Resources (to be updated as we cover more topics)

Lecture Notes
I will  try to regularly post some notes from the lectures.  Not the full set, just some highlights.  And probably not for every lecture

Exercises