|
Activity Number:
|
343
- Contributed Poster Presentations: Section on Bayesian Statistical Science
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 9, 2022 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
|
Abstract #322940
|
|
|
Title:
|
A Bayesian Approach Towards Probability Calibration
|
|
Author(s):
|
Christopher Franck and Adeline Guthrie*
|
|
Companies:
|
Virginia Tech and Virgina Tech
|
|
Keywords:
|
Calibration;
Sports Statistics;
Estimation;
Model Selection;
Bayesian
|
|
Abstract:
|
Properly calibrated probabilities are a key component of the decision-making process. Predicted probabilities are considered well-calibrated when they are consistent with the relative frequency of events they aimed to predict. We develop Bayesian estimation and hypothesis testing-based methodology with a likelihood function specifically suited to the probability calibration problem. This approach allows the evaluations of pundits from areas like sports and entertainment to policy. The aim is to not only assess calibration, but also to rank pundits by level of calibration. We draw connections between our likelihood and a specific generalized non-linear model. We highlight the approach by comparing hockey predictions from competing pundits.
|
Authors who are presenting talks have a * after their name.
