Limit setting and coverage checks in a search for contact interactions
Advisor(s)
Dr. Peter Dong; Illinois Mathematics and Science Academy
Discipline
Physical Science
Start Date
21-4-2021 9:30 AM
End Date
21-4-2021 9:45 AM
Abstract
When analyzing data in particle physics there are two different statistical lenses that people use: Bayesian and frequentist. The core of Bayesian statistics is the subjective belief that a certain result is true or falls within a certain interval while the core of frequentist statistics is the frequency, or amount of times, that the experiments show a result to be true or to fall within a certain interval. Because of how different they are, in order for both Bayesians and frequentists to agree upon the results, there must be a way to reconcile the two--which comes in the form of coverage checks. Throughout the past year I have been finding how good the coverage is for our method of limit setting while searching for contact interactions as a model of quark-lepton compositeness, and with a beta of 0.000558 and beta = 1/lambda^2, the coverage is 99.8% and the median interval of the limit is 33.3 TeV < Lambda < 50.8 TeV.
Limit setting and coverage checks in a search for contact interactions
When analyzing data in particle physics there are two different statistical lenses that people use: Bayesian and frequentist. The core of Bayesian statistics is the subjective belief that a certain result is true or falls within a certain interval while the core of frequentist statistics is the frequency, or amount of times, that the experiments show a result to be true or to fall within a certain interval. Because of how different they are, in order for both Bayesians and frequentists to agree upon the results, there must be a way to reconcile the two--which comes in the form of coverage checks. Throughout the past year I have been finding how good the coverage is for our method of limit setting while searching for contact interactions as a model of quark-lepton compositeness, and with a beta of 0.000558 and beta = 1/lambda^2, the coverage is 99.8% and the median interval of the limit is 33.3 TeV < Lambda < 50.8 TeV.